instruction
stringlengths 999
2.35k
| input
stringlengths 341
562
| output
dict |
|---|---|---|
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['glue', 'blender', 'food_salt', 'toilet', 'chair', 'toothbrush', 'drawing', 'orchid', 'bathroom_counter', 'bookshelf', 'purse', 'alcohol', 'clothes_jacket', 'ceilinglamp', 'foundation', 'pillow', 'bowl', 'dining_room', 'bed', 'curtain', 'lighter', 'doorjamb', 'towel_rack', 'console', 'chessboard', 'cd', 'clothes_scarf', 'instrument_guitar', 'light', 'teeth']
Task:Listen to music
response:
{
"thoughts": I walk to the living room. I find a radio there and switch it on to listen to music.,
"actions": ['[WALK] <home_office>', '[WALK] <stereo>', '[FIND] <stereo>', '[SWITCHON] <stereo>']
}
|
Scene_Objects:['pot', 'tablelamp', 'toilet', 'freezer', 'walllamp', 'bathroom', 'trashcan', 'knifeblock', 'floor', 'ceiling', 'desk', 'bedroom', 'stovefan', 'ceilinglamp', 'plate', 'mat', 'couch', 'wall', 'oven', 'food_food', 'tray', 'food_onion', 'curtain', 'diary', 'kitchen_counter', 'dishrack', 'sauce_pan', 'cupboard', 'spoon', 'home_office']
Task:Cook some food
response:
| {
"thoughts": "I pull a sauce pan from the cupboard. I turn the stove on. I take my ingredients from the pantry and fridge. I put the ingredients in the pan and I mix them with a spoon. I cook until the food is finished and transfer my food to a plate.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <food_food>",
"[FIND] <food_food>",
"[FIND] <pot>",
"[GRAB] <pot>",
"[FIND] <spoon>",
"[GRAB] <spoon>",
"[FIND] <sauce_pan>",
"[PUTBACK] <spoon> <sauce_pan>",
"[FIND] <freezer>",
"[OPEN] <freezer>",
"[FIND] <plate>",
"[PUTBACK] <pot> <plate>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sponge', 'microwave', 'cpuscreen', 'floor', 'ceiling', 'phone', 'character', 'folder', 'desk', 'alcohol', 'stovefan', 'plate', 'mousepad', 'wall', 'oven', 'food_food', 'check', 'tray', 'doorjamb', 'cards', 'mouse', 'kitchen_counter', 'shoes', 'cupboard', 'clothes_dress', 'sink', 'clothes_scarf', 'light', 'clothes_shirt', 'teeth']
Task:Pick up phone
response:
{
"thoughts": walk over and pick up the phone,
"actions": ['[WALK] <phone>', '[GRAB] <phone>']
}
|
Scene_Objects:['food_ice_cream', 'chair', 'walllamp', 'trashcan', 'bills', 'knife', 'floor', 'desk', 'mail', 'button', 'toaster', 'plate', 'faucet', 'shower', 'foundation', 'couch', 'bowl', 'wall', 'nightstand', 'oven', 'bed', 'clothes_gloves', 'novel', 'coffe_maker', 'check', 'tray', 'sink', 'table', 'dishwasher', 'shoe_rack']
Task:Wash dishes with dishwasher
response:
| {
"thoughts": "In the kitchen, open up the dishwasher. Take plates from the sink and put them in the dishwasher one by one. Close the dishwasher. Program it. Push start.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <dishwasher>",
"[FIND] <dishwasher>",
"[OPEN] <dishwasher>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[PUTBACK] <plate> <dishwasher>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[PUTBACK] <plate> <dishwasher>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[PUTBACK] <plate> <dishwasher>",
"[CLOSE] <dishwasher>",
"[SWITCHON] <dishwasher>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['toilet', 'freezer', 'walllamp', 'cat', 'ceilingfan', 'drawing', 'cpuscreen', 'clothes_underwear', 'longboard', 'floor', 'ceiling', 'keyboard', 'ceilinglamp', 'tvstand', 'balanceball', 'window', 'oil', 'couch', 'pillow', 'wall', 'photoframe', 'food_food', 'television', 'ice', 'curtain', 'after_shave', 'door', 'table', 'dishwasher', 'vase']
Task:Watch TV
response:
{
"thoughts": Sit on couch and watch the tv,
"actions": ['[WALK] <home_office>', '[WALK] <television>', '[FIND] <television>', '[SWITCHON] <television>', '[FIND] <couch>', '[SIT] <couch>', '[TURNTO] <television>', '[WATCH] <television>']
}
|
Scene_Objects:['freezer', 'blow_dryer', 'ceilingfan', 'trashcan', 'toothbrush', 'drawing', 'knife', 'hanger', 'floor', 'ceiling', 'bookshelf', 'stovefan', 'ceilinglamp', 'tvstand', 'balanceball', 'window', 'couch', 'wall', 'coffe_maker', 'television', 'food_donut', 'curtain', 'coffee_pot', 'controller', 'sink', 'bathroom_cabinet', 'table', 'soap', 'teeth', 'shoe_rack']
Task:Watch TV
response:
| {
"thoughts": "I walk to the living room. I grab the remote control. I walk to the sofa. I sit down, and press the on button on the remote.",
"actions": [
"[WALK] <home_office>",
"[WALK] <remote_control>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[WALK] <couch>",
"[SIT] <couch>",
"[TOUCH] <remote_control>",
"[FIND] <television>",
"[SWITCHON] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'filing_cabinet', 'trashcan', 'drawing', 'duster', 'wallshelf', 'floor', 'ceiling', 'bookshelf', 'keyboard', 'stovefan', 'mat', 'tvstand', 'picture', 'window', 'mousepad', 'couch', 'food_peanut_butter', 'pillow', 'wall', 'powersocket', 'photoframe', 'food_food', 'curtain', 'doorjamb', 'sink', 'bathroom_cabinet', 'light', 'home_office', 'scrabble']
Task:Dust
response:
{
"thoughts": Using the duster, lightly dust all hanging pictures on walls in living room. Dust all around surface of picture frame and front of pictures, moving duster back and forth, side to side, until all dust is gone.,
"actions": ['[WALK] <home_office>', '[WALK] <filing_cabinet>', '[OPEN] <filing_cabinet>', '[FIND] <duster>', '[GRAB] <duster>', '[WALK] <wall>', '[WALK] <picture>', '[WIPE] <picture>', '[WALK] <picture>', '[WIPE] <picture>', '[WALK] <picture>', '[WIPE] <picture>', '[WALK] <filing_cabinet>', '[PUTBACK] <duster> <filing_cabinet>', '[CLOSE] <filing_cabinet>']
}
|
Scene_Objects:['sponge', 'pot', 'fork', 'beer', 'walllamp', 'bathroom', 'orchid', 'kitchen_cabinet', 'floor', 'phone', 'toaster', 'mat', 'plate', 'tvstand', 'window', 'dish_soap', 'couch', 'bowl', 'wall', 'oven', 'bench', 'coffe_maker', 'tray', 'rag', 'console', 'kitchen_counter', 'dishrack', 'hair', 'maindoor', 'dishwasher']
Task:Wash dishes with dishwasher
response:
| {
"thoughts": "I went to the kitchen and collected the dishes and put them into the dish washer and wiped them with towel and put back into cabinet..",
"actions": [
"[WALK] <dining_room>",
"[WALK] <kitchen_cabinet>",
"[FIND] <kitchen_cabinet>",
"[OPEN] <kitchen_cabinet>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[FIND] <dishwasher>",
"[OPEN] <dishwasher>",
"[PUTBACK] <plate> <dishwasher>",
"[FIND] <dish_soap>",
"[GRAB] <dish_soap>",
"[PUTBACK] <dish_soap> <dishwasher>",
"[CLOSE] <dishwasher>",
"[SWITCHON] <dishwasher>",
"[GRAB] <plate>",
"[WIPE] <plate>",
"[FIND] <dishrack>",
"[GRAB] <dishrack>",
"[PUTOBJBACK] <dishrack>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['spectacles', 'tablelamp', 'beer', 'chair', 'walllamp', 'bathroom', 'wallshelf', 'floor', 'ceiling', 'ceilinglamp', 'mat', 'stereo', 'couch', 'pillow', 'bowl', 'wall', 'nightstand', 'powersocket', 'cup', 'toilet_paper', 'television', 'after_shave', 'rag', 'doorjamb', 'coffee_pot', 'dishrack', 'instrument_guitar', 'table', 'soap', 'laser_pointer']
Task:Relax on sofa
response:
{
"thoughts": walk to living room, walk to sofa, sit in sofa,
"actions": ['[WALK] <home_office>', '[WALK] <couch>', '[SIT] <couch>']
}
|
Scene_Objects:['juice', 'detergent', 'knifeblock', 'cpuscreen', 'longboard', 'folder', 'character', 'bookshelf', 'table_cloth', 'toaster', 'stereo', 'picture', 'food_orange', 'nightstand', 'wall_clock', 'tape', 'light_bulb', 'coffe_maker', 'ice', 'tray', 'curtain', 'doorjamb', 'cd', 'shoes', 'dresser', 'door', 'sink', 'light', 'dishwasher', 'teeth']
Task:Listen to music
response:
| {
"thoughts": "I walk to the living room. I find a radio there and switch it on to listen to music.",
"actions": [
"[WALK] <home_office>",
"[WALK] <stereo>",
"[FIND] <stereo>",
"[SWITCHON] <stereo>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'microwave', 'toilet', 'chair', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'button', 'mail', 'ceilinglamp', 'deck_of_cards', 'balanceball', 'window', 'mousepad', 'wall', 'nightstand', 'photoframe', 'novel', 'doorjamb', 'mouse', 'light', 'dishwasher']
Task:Write an email
response:
{
"thoughts": Switch on computer, Use keybord and mouse to open e-mail provider, type requested e-mail, click send, close e-mail provider, turn off computer,
"actions": ['[WALK] <home_office>', '[WALK] <computer>', '[FIND] <computer>', '[SWITCHON] <computer>', '[FIND] <mouse>', '[FIND] <keyboard>', '[FIND] <mail>', '[OPEN] <mail>', '[TYPE] <keyboard>', '[FIND] <button>', '[PUSH] <button>', '[CLOSE] <mail>', '[SWITCHOFF] <computer>']
}
|
Scene_Objects:['water_glass', 'filing_cabinet', 'food_salt', 'microwave', 'freezer', 'walllamp', 'ground_coffee', 'knifeblock', 'needle', 'floor', 'mail', 'plate', 'faucet', 'window', 'pillow', 'nightstand', 'basket_for_clothes', 'cup', 'photoframe', 'bed', 'coffe_maker', 'coffee_filter', 'coin', 'water', 'after_shave', 'dresser', 'hairbrush', 'maindoor', 'table', 'shoe_rack']
Task:Drink
response:
| {
"thoughts": "walk to kitchen, look at the water glass, grab that glass, find the water cane, open the water cane tap, drink water",
"actions": [
"[WALK] <dining_room>",
"[WALK] <water_glass>",
"[FIND] <water_glass>",
"[TURNTO] <water_glass>",
"[LOOKAT] <water_glass>",
"[GRAB] <water_glass>",
"[FIND] <water>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sheets', 'filing_cabinet', 'microwave', 'ground_coffee', 'knifeblock', 'wallshelf', 'ceiling', 'phone', 'character', 'desk', 'piano_bench', 'stovefan', 'toaster', 'couch', 'clothes_hat', 'pillow', 'bench', 'oven', 'tape', 'cutting_board', 'doorjamb', 'cards', 'mouse', 'dishrack', 'cupboard', 'hairdryer', 'sink', 'clothes_scarf', 'table', 'dishwasher']
Task:Pick up phone
response:
{
"thoughts": i go find the phone. i pick up the phone,
"actions": ['[WALK] <home_office>', '[WALK] <phone>', '[FIND] <phone>', '[GRAB] <phone>']
}
|
Scene_Objects:['tablelamp', 'toilet', 'chair', 'walllamp', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'ceilinglamp', 'balanceball', 'mousepad', 'oil', 'wall', 'nightstand', 'photoframe', 'doorjamb', 'mouse', 'dresser', 'cupboard', 'clothes_dress', 'door', 'light']
Task:Browse internet
response:
| {
"thoughts": "I want to surf the internet and look around at funny pictures.",
"actions": [
"[WALK] <home_office>",
"[WALK] <light>",
"[FIND] <light>",
"[SWITCHON] <light>",
"[FIND] <keyboard>",
"[GRAB] <keyboard>",
"[FIND] <desk>",
"[OPEN] <desk>",
"[PUTBACK] <keyboard> <desk>",
"[FIND] <mouse>",
"[GRAB] <mouse>",
"[FIND] <mousepad>",
"[PUTBACK] <mouse> <mousepad>",
"[PLUGIN] <keyboard>",
"[PLUGIN] <mouse>",
"[FIND] <computer>",
"[SWITCHON] <computer>",
"[FIND] <chair>",
"[SIT] <chair>",
"[TURNTO] <computer>",
"[LOOKAT] <computer>",
"[GRAB] <mouse>",
"[TOUCH] <mouse>",
"[TYPE] <keyboard>",
"[TURNTO] <computer>",
"[LOOKAT] <computer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['bathroom', 'detergent', 'cat', 'knife', 'food_egg', 'character', 'stamp', 'bedroom', 'desk', 'mat', 'ceilinglamp', 'shower', 'faucet', 'stereo', 'window', 'oil', 'mousepad', 'clothes_hat', 'towel', 'pillow', 'dining_room', 'powersocket', 'coin', 'console', 'towel_rack', 'dishrack', 'cupboard', 'maindoor', 'sink', 'bag']
Task:Keep cats out of room
response:
{
"thoughts": Stand in the doorway and create a loud noise so the cats will stay out.,
"actions": ['[WALK] <bedroom>', '[WALK] <cat>', '[FIND] <cat>', '[TURNTO] <cat>', '[LOOKAT] <cat>', '[FIND] <cat>', '[TURNTO] <cat>', '[LOOKAT] <cat>']
}
|
Scene_Objects:['tablelamp', 'toilet', 'chair', 'nail_polish', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'clothes_underwear', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'mail', 'ceilinglamp', 'balanceball', 'mousepad', 'dustpan', 'towel', 'food_kiwi', 'wall', 'nightstand', 'novel', 'doorjamb', 'mouse', 'home_office', 'light']
Task:Write an email
response:
| {
"thoughts": "go to the office and use the computer to write and email",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>",
"[FIND] <mail>",
"[OPEN] <mail>",
"[FIND] <keyboard>",
"[TYPE] <keyboard>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['scissors', 'walllamp', 'bathroom', 'knifeblock', 'wallshelf', 'floor', 'ceiling', 'bookshelf', 'keyboard', 'stovefan', 'ceilinglamp', 'tvstand', 'shower', 'plate', 'window', 'toaster', 'clothes_pants', 'broom', 'powersocket', 'bench', 'bed', 'check', 'food_food', 'coffe_maker', 'after_shave', 'dresser', 'hair', 'door', 'sink', 'clothes_shirt']
Task:Get dressed
response:
{
"thoughts": Get some clothes together that make an appropriate outfit. Put them on,
"actions": ['[WALK] <bedroom>', '[WALK] <dresser>', '[FIND] <dresser>', '[OPEN] <dresser>', '[FIND] <clothes_pants>', '[GRAB] <clothes_pants>', '[PUTON] <clothes_pants>', '[FIND] <clothes_shirt>', '[GRAB] <clothes_shirt>', '[PUTON] <clothes_shirt>', '[CLOSE] <dresser>']
}
|
Scene_Objects:['spectacles', 'tea', 'ground_coffee', 'walllamp', 'hanger', 'longboard', 'floor', 'ceiling', 'button', 'toaster', 'faucet', 'mousepad', 'candle', 'wall', 'nightstand', 'powersocket', 'keys', 'novel', 'coffe_maker', 'clothes_gloves', 'coffee_filter', 'water', 'ice', 'tray', 'food_onion', 'cupboard', 'door', 'light', 'dishwasher', 'soap']
Task:Make coffee
response:
| {
"thoughts": "Open coffee maker. Put filter in main basket, put ground coffee in filter, fill reservoir with water, turn on coffee maker.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <coffe_maker>",
"[FIND] <coffe_maker>",
"[OPEN] <coffe_maker>",
"[FIND] <coffee_filter>",
"[GRAB] <coffee_filter>",
"[PUTBACK] <coffee_filter> <coffe_maker>",
"[FIND] <ground_coffee>",
"[GRAB] <ground_coffee>",
"[PUTBACK] <ground_coffee> <coffe_maker>",
"[WALK] <water>",
"[FIND] <water>",
"[GRAB] <water>",
"[POUR] <water> <coffe_maker>",
"[CLOSE] <coffe_maker>",
"[SWITCHON] <coffe_maker>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['fork', 'tablelamp', 'crayon', 'microwave', 'drawing', 'bills', 'cpuscreen', 'longboard', 'floor', 'bookshelf', 'computer', 'button', 'table_cloth', 'toaster', 'ceilinglamp', 'faucet', 'window', 'oil', 'couch', 'food_jam', 'food_orange', 'wall', 'bench', 'mouthwash', 'television', 'cupboard', 'table', 'clothes_skirt', 'light', 'tooth_paste']
Task:Relax on sofa
response:
{
"thoughts": walk into living room, look at the sofa which one is available, sit on the sofa, now take relax on the sofa,
"actions": ['[WALK] <home_office>', '[WALK] <couch>', '[FIND] <couch>', '[TURNTO] <couch>', '[LOOKAT] <couch>', '[SIT] <couch>']
}
|
Scene_Objects:['toilet', 'ceilingfan', 'drawing', 'floor', 'ceiling', 'keyboard', 'ceilinglamp', 'laptop', 'faucet', 'tvstand', 'window', 'couch', 'microphone', 'towel', 'pillow', 'wall', 'oven', 'powersocket', 'photoframe', 'novel', 'food_food', 'television', 'tray', 'mouse', 'dresser', 'sink', 'clothes_scarf', 'home_office', 'table', 'clothes_skirt']
Task:Relax on sofa
response:
| {
"thoughts": "I walk to the living room. I see the couch. I sit down on the couch, then lie down and put my feet up.",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[SIT] <couch>",
"[LIE] <couch>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'tablelamp', 'food_salt', 'microwave', 'toilet', 'walllamp', 'drawing', 'needle', 'hanger', 'bathroom_counter', 'ceiling', 'bedroom', 'desk', 'computer', 'bookmark', 'stovefan', 'plate', 'tvstand', 'dish_soap', 'closetdrawer', 'dining_room', 'wall', 'wall_clock', 'oven', 'powersocket', 'curtain', 'kitchen_counter', 'sink', 'bathroom_cabinet', 'dishwasher']
Task:Wash dishes with dishwasher
response:
{
"thoughts": Walk to the kitchen. Open the dishwasher. Grab plate and put it in the dishwasher. Grab dish soap and put it in the dishwasher. Switch on the dishwasher.,
"actions": ['[WALK] <dining_room>', '[WALK] <dishwasher>', '[FIND] <dishwasher>', '[OPEN] <dishwasher>', '[FIND] <plate>', '[GRAB] <plate>', '[PUTBACK] <plate> <dishwasher>', '[FIND] <dish_soap>', '[GRAB] <dish_soap>', '[PUTBACK] <dish_soap> <dishwasher>', '[CLOSE] <dishwasher>', '[SWITCHON] <dishwasher>']
}
|
Scene_Objects:['tablelamp', 'toilet', 'chair', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'bedroom', 'keyboard', 'bookshelf', 'computer', 'alcohol', 'ceilinglamp', 'balanceball', 'mousepad', 'wall', 'nightstand', 'food_cheese', 'toothbrush_holder', 'doorjamb', 'towel_rack', 'mouse', 'cupboard', 'bathroom_cabinet', 'light', 'home_office']
Task:Work
response:
| {
"thoughts": "Type on a computer in the home office.",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>",
"[FIND] <keyboard>",
"[TYPE] <keyboard>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sheets', 'food_ice_cream', 'measuring_cup', 'knife', 'orchid', 'cpuscreen', 'bathroom_counter', 'ceiling', 'phone', 'alcohol', 'computer', 'table_cloth', 'mat', 'deck_of_cards', 'stereo', 'window', 'pillow', 'powersocket', 'cup', 'newspaper', 'check', 'ice', 'after_shave', 'lighter', 'diary', 'cd', 'cupboard', 'maindoor', 'colander', 'board_game']
Task:Listen to music
response:
{
"thoughts": Find radio, turn it on,
"actions": ['[WALK] <home_office>', '[WALK] <stereo>', '[FIND] <stereo>', '[SWITCHON] <stereo>']
}
|
Scene_Objects:['pot', 'scissors', 'freezer', 'bathroom', 'orchid', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'bookshelf', 'piano_bench', 'ceilinglamp', 'tvstand', 'faucet', 'clothes_socks', 'pillow', 'wall', 'nightstand', 'bed', 'newspaper', 'food_onion', 'doorjamb', 'towel_rack', 'shoes', 'dresser', 'hair', 'hairdryer', 'maindoor', 'spoon', 'table']
Task:Put groceries in Fridge
response:
| {
"thoughts": "In the kitchen where I left the bags of groceries, I grab cheese from the grocery bag, I walk over to the refrigerator, open it up, neatly put cheese in the fridge, Close the fridge",
"actions": [
"[WALK] <dining_room>",
"[WALK] <food_food>",
"[FIND] <food_food>",
"[GRAB] <food_food>",
"[WALK] <freezer>",
"[OPEN] <freezer>",
"[PUTIN] <food_food> <freezer>",
"[CLOSE] <freezer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'crayon', 'freezer', 'bathroom', 'trashcan', 'dough', 'knifeblock', 'orchid', 'floor', 'ceiling', 'clothes_jacket', 'brush', 'ceilinglamp', 'mat', 'balanceball', 'clothes_socks', 'couch', 'microphone', 'form', 'wall', 'food_food', 'check', 'oven_mitts', 'wine', 'mouse', 'cupboard', 'food_steak', 'spoon', 'table', 'tooth_paste']
Task:Cook some food
response:
{
"thoughts": Go kitchen and make something for eat,
"actions": ['[WALK] <dining_room>', '[WALK] <freezer>', '[OPEN] <freezer>', '[FIND] <food_food>']
}
|
Scene_Objects:['water_glass', 'microwave', 'needle', 'kitchen_cabinet', 'desk', 'alcohol', 'mat', 'ceilinglamp', 'balanceball', 'window', 'couch', 'clothes_pants', 'dining_room', 'nightstand', 'wall', 'wall_clock', 'oven', 'toilet_paper', 'mouthwash', 'water', 'ice', 'tray', 'rag', 'doorjamb', 'towel_rack', 'coffee_pot', 'pencil', 'door', 'dishwasher', 'vase']
Task:Drink
response:
| {
"thoughts": "Get a glass of water to drink",
"actions": [
"[WALK] <dining_room>",
"[WALK] <kitchen_cabinet>",
"[FIND] <kitchen_cabinet>",
"[FIND] <kitchen_cabinet>",
"[FIND] <water>",
"[GRAB] <water>",
"[FIND] <water_glass>",
"[POUR] <water> <water_glass>",
"[GRAB] <water_glass>",
"[DRINK] <water_glass>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['microwave', 'chair', 'walllamp', 'food_carrot', 'orchid', 'knife', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'food_bacon', 'ceilinglamp', 'mat', 'tvstand', 'stereo', 'mousepad', 'couch', 'wall', 'powersocket', 'toy', 'photoframe', 'mouse', 'controller', 'door', 'table', 'light']
Task:Browse internet
response:
{
"thoughts": turn on the computer and browse the internet,
"actions": ['[WALK] <home_office>', '[WALK] <computer>', '[FIND] <computer>', '[SWITCHON] <computer>']
}
|
Scene_Objects:['pot', 'filing_cabinet', 'food_pizza', 'food_salt', 'jelly', 'band-aids', 'drawing', 'orchid', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'bookshelf', 'keyboard', 'computer', 'mat', 'ceilinglamp', 'broom', 'wall', 'powersocket', 'doorjamb', 'lighter', 'door', 'sink', 'light', 'table', 'laser_pointer', 'clothes_shirt', 'creditcard']
Task:Turn on light
response:
| {
"thoughts": "I go into my room, find the light, then turn on the lights.",
"actions": [
"[WALK] <bedroom>",
"[WALK] <light>",
"[FIND] <light>",
"[WALK] <light>",
"[SWITCHON] <light>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['blender', 'toilet', 'chair', 'walllamp', 'detergent', 'food_carrot', 'orchid', 'cpuscreen', 'hanger', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'computer', 'bookmark', 'mat', 'window', 'mousepad', 'broom', 'wall', 'wall_clock', 'nightstand', 'novel', 'coin', 'lighter', 'centerpiece', 'mouse', 'dresser']
Task:Read book
response:
{
"thoughts": find a book to read. sit down. read the book.,
"actions": ['[WALK] <home_office>', '[WALK] <novel>', '[FIND] <novel>', '[GRAB] <novel>', '[WALK] <chair>', '[SIT] <chair>', '[READ] <novel>']
}
|
Scene_Objects:['water_glass', 'pot', 'food_salt', 'mop', 'freezer', 'walllamp', 'detergent', 'knifeblock', 'floor', 'ceiling', 'purse', 'button', 'piano_bench', 'food_bacon', 'stovefan', 'ceilinglamp', 'tvstand', 'faucet', 'form', 'wall', 'bench', 'food_food', 'television', 'water', 'kitchen_counter', 'cupboard', 'food_steak', 'sink', 'maindoor', 'table']
Task:Drink
response:
| {
"thoughts": "I need to go to Kitchen. Get a cup. Fill cup. Drink",
"actions": [
"[WALK] <dining_room>",
"[WALK] <cupboard>",
"[FIND] <cupboard>",
"[OPEN] <cupboard>",
"[FIND] <water_glass>",
"[GRAB] <water_glass>",
"[WALK] <freezer>",
"[WALK] <water>",
"[FIND] <water>",
"[GRAB] <water>",
"[POUR] <water> <water_glass>",
"[DRINK] <water_glass>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['fork', 'filing_cabinet', 'chair', 'drawing', 'bills', 'knife', 'floor', 'ceilinglamp', 'stereo', 'window', 'couch', 'food_jam', 'form', 'towel', 'dining_room', 'wall', 'basket_for_clothes', 'photoframe', 'novel', 'food_food', 'check', 'television', 'oven_mitts', 'razor', 'tray', 'pencil', 'dresser', 'pajamas', 'table', 'soap']
Task:Read book
response:
{
"thoughts": Sit and read,
"actions": ['[WALK] <home_office>', '[WALK] <filing_cabinet>', '[OPEN] <filing_cabinet>', '[FIND] <novel>', '[GRAB] <novel>', '[CLOSE] <filing_cabinet>', '[WALK] <couch>', '[SIT] <couch>', '[READ] <novel>']
}
|
Scene_Objects:['tablelamp', 'microwave', 'board_game', 'walllamp', 'bathroom', 'ceilingfan', 'drawing', 'longboard', 'phone', 'ceiling', 'dvd_player', 'bedroom', 'tea_bag', 'stamp', 'keyboard', 'stovefan', 'plate', 'picture', 'form', 'pillow', 'oven', 'television', 'centerpiece', 'shoes', 'cd', 'bag', 'dishwasher', 'soap', 'tooth_paste', 'creditcard']
Task:Pet cat
response:
| {
"thoughts": "Find the cat. Pet the cat.",
"actions": [
"[WALK] <home_office>",
"[WALK] <cat>",
"[FIND] <cat>",
"[TOUCH] <cat>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['fork', 'chair', 'walllamp', 'drawing', 'wallshelf', 'hanger', 'floor', 'ceiling', 'desk', 'remote_control', 'computer', 'mat', 'tvstand', 'ceilinglamp', 'closetdrawer', 'couch', 'form', 'pillow', 'wall', 'pasta', 'powersocket', 'photoframe', 'television', 'food_donut', 'curtain', 'doorjamb', 'chessboard', 'dresser', 'laundry_detergent', 'table']
Task:Watch TV
response:
{
"thoughts": walk to living room, find couch, sit on couch, find remote control , turn on tv by pressing button,
"actions": ['[WALK] <home_office>', '[WALK] <couch>', '[FIND] <couch>', '[SIT] <couch>', '[FIND] <remote_control>', '[FIND] <television>', '[TURNTO] <television>']
}
|
Scene_Objects:['filing_cabinet', 'microwave', 'ceilingfan', 'drawing', 'knifeblock', 'knife', 'floor', 'ceiling', 'remote_control', 'ceilinglamp', 'tvstand', 'balanceball', 'mat', 'window', 'stereo', 'food_noodles', 'couch', 'dustpan', 'form', 'dining_room', 'wall', 'powersocket', 'oven', 'check', 'television', 'curtain', 'kitchen_counter', 'cd', 'table', 'vase']
Task:Watch TV
response:
| {
"thoughts": "Go to the sofa and watch TV",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[SIT] <couch>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[FIND] <television>",
"[TURNTO] <television>",
"[WATCH] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'chair', 'walllamp', 'bathroom', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'character', 'desk', 'keyboard', 'computer', 'ceilinglamp', 'tvstand', 'window', 'mousepad', 'wall', 'wall_clock', 'powersocket', 'oven', 'check', 'television', 'oven_mitts', 'doorjamb', 'mouse', 'kitchen_counter', 'sink', 'napkin', 'table', 'shoe_rack']
Task:Browse internet
response:
{
"thoughts": I will walk to the computer at the desk in my home office, sit in the chair, turn on the computer, open a web browser, type in the address, and view the computer.,
"actions": ['[WALK] <home_office>', '[WALK] <desk>', '[FIND] <chair>', '[SIT] <chair>', '[FIND] <computer>', '[SWITCHON] <computer>', '[FIND] <mouse>', '[GRAB] <mouse>', '[TOUCH] <mouse>', '[FIND] <keyboard>', '[GRAB] <keyboard>', '[TYPE] <keyboard>', '[TURNTO] <computer>', '[LOOKAT] <computer>']
}
|
Scene_Objects:['pot', 'filing_cabinet', 'freezer', 'bathroom', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'character', 'desk', 'bookshelf', 'keyboard', 'computer', 'clothes_jacket', 'table_cloth', 'mat', 'food_kiwi', 'wall', 'powersocket', 'clothes_gloves', 'food_food', 'coffe_maker', 'doorjamb', 'coffee_pot', 'headset', 'door', 'hairdryer', 'light', 'tooth_paste']
Task:Turn on light
response:
| {
"thoughts": "I walk towards to the bedroom, and I open the door of the bedroom, and I switch on the lights inside the bedroom, the switch is beside the door.",
"actions": [
"[WALK] <bedroom>",
"[WALK] <light>",
"[FIND] <light>",
"[SWITCHON] <light>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'toilet', 'bathroom', 'orchid', 'cpuscreen', 'wallshelf', 'floor', 'phone', 'character', 'bedroom', 'toaster', 'ceilinglamp', 'stovefan', 'wall_clock', 'powersocket', 'bed', 'food_food', 'television', 'wooden_spoon', 'kitchen_counter', 'dresser', 'cupboard', 'door', 'sink', 'hairdryer', 'home_office', 'light', 'clothes_skirt', 'bathtub', 'bathroom_cabinet']
Task:Read book
response:
{
"thoughts": pick up your book and read it,
"actions": ['[WALK] <bedroom>', '[WALK] <novel>', '[GRAB] <novel>', '[READ] <novel>']
}
|
Scene_Objects:['toilet', 'walllamp', 'drawing', 'orchid', 'knifeblock', 'bathroom_counter', 'floor', 'ceiling', 'table_cloth', 'mat', 'ceilinglamp', 'shower', 'plate', 'wall', 'nightstand', 'powersocket', 'food_cheese', 'bed', 'coffe_maker', 'tray', 'curtain', 'towel_rack', 'dresser', 'controller', 'clothes_scarf', 'home_office', 'bathroom_cabinet', 'board_game', 'scrabble', 'teeth']
Task:Take shower
response:
| {
"thoughts": "walk to bathroom , find shower , take a shower",
"actions": [
"[WALK] <bathroom>",
"[WALK] <shower>",
"[FIND] <shower>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'toilet', 'chair', 'walllamp', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'ceilinglamp', 'balanceball', 'mousepad', 'oil', 'wall', 'nightstand', 'photoframe', 'doorjamb', 'mouse', 'dresser', 'cupboard', 'clothes_dress', 'door', 'light']
Task:Browse internet
response:
{
"thoughts": I want to surf the internet and look around at funny pictures.,
"actions": ['[WALK] <home_office>', '[WALK] <light>', '[FIND] <light>', '[SWITCHON] <light>', '[FIND] <keyboard>', '[GRAB] <keyboard>', '[FIND] <desk>', '[OPEN] <desk>', '[PUTBACK] <keyboard> <desk>', '[FIND] <mouse>', '[GRAB] <mouse>', '[FIND] <mousepad>', '[PUTBACK] <mouse> <mousepad>', '[PLUGIN] <keyboard>', '[PLUGIN] <mouse>', '[FIND] <computer>', '[SWITCHON] <computer>', '[FIND] <chair>', '[SIT] <chair>', '[TURNTO] <computer>', '[LOOKAT] <computer>', '[GRAB] <mouse>', '[TOUCH] <mouse>', '[TYPE] <keyboard>', '[TURNTO] <computer>', '[LOOKAT] <computer>']
}
|
Scene_Objects:['pot', 'measuring_cup', 'walllamp', 'bathroom', 'trashcan', 'ceilingfan', 'orchid', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'shampoo', 'character', 'toaster', 'mat', 'dish_soap', 'couch', 'wall', 'nightstand', 'oven', 'powersocket', 'bench', 'photoframe', 'coffe_maker', 'tray', 'doorjamb', 'shaving_cream', 'cupboard', 'food_steak', 'dishwasher']
Task:Wash dishes with dishwasher
response:
| {
"thoughts": "I go to the dishwasher and put dish soap. I close the dishwasher and turn it on.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <dishwasher>",
"[FIND] <dishwasher>",
"[OPEN] <dishwasher>",
"[FIND] <dish_soap>",
"[GRAB] <dish_soap>",
"[PUTBACK] <dish_soap> <dishwasher>",
"[GRAB] <dish_soap>",
"[PUTBACK] <dish_soap> <dishwasher>",
"[CLOSE] <dishwasher>",
"[SWITCHON] <dishwasher>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'filing_cabinet', 'scissors', 'mop', 'microwave', 'toilet', 'walllamp', 'drawing', 'orchid', 'cpuscreen', 'floor', 'desk', 'bookshelf', 'tvstand', 'shower', 'stereo', 'mousepad', 'wall', 'powersocket', 'photoframe', 'bed', 'food_food', 'television', 'curtain', 'doorjamb', 'coffee_pot', 'chessboard', 'laundry_detergent', 'door', 'board_game']
Task:Go to toilet
response:
{
"thoughts": walk to bathroom , run to toilet,
"actions": ['[WALK] <bathroom>', '[RUN] <toilet>']
}
|
Scene_Objects:['chair', 'band-aids', 'trashcan', 'dough', 'knifeblock', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'shampoo', 'dvd_player', 'desk', 'bookshelf', 'keyboard', 'computer', 'stovefan', 'mat', 'tvstand', 'shower', 'mousepad', 'food_orange', 'wall', 'oven', 'keys', 'mouse', 'shoes', 'hair', 'dishwasher', 'vase', 'coffee']
Task:Read book
response:
| {
"thoughts": "find a book to read. sit down. read the book.",
"actions": [
"[WALK] <home_office>",
"[WALK] <novel>",
"[FIND] <novel>",
"[GRAB] <novel>",
"[WALK] <chair>",
"[SIT] <chair>",
"[READ] <novel>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['mop', 'toilet', 'freezer', 'food_carrot', 'cpuscreen', 'bathroom_counter', 'floor', 'ceiling', 'piano_bench', 'stovefan', 'mat', 'faucet', 'wall', 'wall_clock', 'oven', 'cutting_board', 'toilet_paper', 'video_game_controller', 'television', 'tray', 'doorjamb', 'towel_rack', 'coffee_pot', 'kitchen_counter', 'dresser', 'conditioner', 'sink', 'bathroom_cabinet', 'light', 'teeth']
Task:Wash teeth
response:
{
"thoughts": Walk to bathroom. Find mouthwash or turn on water. Put mouthwash or water in mouth. Swish around for 30 seconds. Spit mouthwash out.,
"actions": ['[WALK] <bathroom>', '[WALK] <faucet>', '[FIND] <faucet>', '[SWITCHON] <faucet>', '[FIND] <teeth>', '[RINSE] <teeth>']
}
|
Scene_Objects:['glue', 'pot', 'scissors', 'walllamp', 'detergent', 'drawing', 'clothes_underwear', 'floor', 'phone', 'character', 'button', 'toaster', 'plate', 'tvstand', 'dish_soap', 'oil', 'candle', 'wall', 'oven', 'tape', 'box', 'coffe_maker', 'food_food', 'check', 'tray', 'centerpiece', 'controller', 'light', 'dishwasher', 'soap']
Task:Wash dishes with dishwasher
response:
| {
"thoughts": "walk into kitchen, find dishwasher, put plates in dishwasher, put soap in dishwasher, turn on dishwasher",
"actions": [
"[WALK] <dining_room>",
"[WALK] <dishwasher>",
"[FIND] <dishwasher>",
"[OPEN] <dishwasher>",
"[FIND] <dish_soap>",
"[GRAB] <dish_soap>",
"[PUTBACK] <dish_soap> <dishwasher>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[PUTBACK] <plate> <dishwasher>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[PUTBACK] <plate> <dishwasher>",
"[CLOSE] <dishwasher>",
"[SWITCHON] <dishwasher>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'chair', 'walllamp', 'ground_coffee', 'detergent', 'trashcan', 'orchid', 'drawing', 'bedroom', 'tea_bag', 'bookshelf', 'envelope', 'alcohol', 'computer', 'ceilinglamp', 'mat', 'shower', 'deck_of_cards', 'food_noodles', 'microphone', 'bowl', 'wall', 'bench', 'powersocket', 'photoframe', 'television', 'dresser', 'napkin', 'table', 'clothes_shirt']
Task:Pet cat
response:
{
"thoughts": First I need to find the cat in the living room. Then I walk over to the cat and pet it gently.,
"actions": ['[WALK] <home_office>', '[WALK] <cat>', '[FIND] <cat>', '[WALK] <cat>', '[TOUCH] <cat>']
}
|
Scene_Objects:['toilet', 'chair', 'freezer', 'hanger', 'floor', 'ceiling', 'phone', 'dvd_player', 'character', 'desk', 'bedroom', 'bookshelf', 'envelope', 'computer', 'ceilinglamp', 'stereo', 'wall', 'pasta', 'cup', 'tray', 'doorjamb', 'dresser', 'cupboard', 'hair', 'clothes_dress', 'door', 'clothes_scarf', 'maindoor', 'light', 'dishwasher']
Task:Turn on light
response:
| {
"thoughts": "Walk to the Living Room. Switch on 1 light. Walk to dining room. Switch on 1 light. Walk to Entrance hall. Switch on 1 light.",
"actions": [
"[WALK] <home_office>",
"[WALK] <light>",
"[FIND] <light>",
"[SWITCHON] <light>",
"[WALK] <dining_room>",
"[WALK] <home_office>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_pizza', 'freezer', 'trashcan', 'drawing', 'bathroom_counter', 'ceiling', 'phone', 'thread', 'computer', 'stovefan', 'faucet', 'shower', 'food_jam', 'dining_room', 'wall', 'powersocket', 'tape', 'coffe_maker', 'food_food', 'mirror', 'doorjamb', 'mouse', 'paper_towel', 'dresser', 'cupboard', 'sink', 'clothes_scarf', 'bathroom_cabinet', 'home_office', 'dirt']
Task:Clean mirror
response:
{
"thoughts": Spray Windex or glass and mirror cleaner onto mirror. Use paper towel to wipe off cleaner until mirror is dry and has no streaks.,
"actions": ['[WALK] <bathroom>', '[WALK] <bathroom_cabinet>', '[FIND] <bathroom_cabinet>', '[OPEN] <bathroom_cabinet>', '[FIND] <paper_towel>', '[GRAB] <paper_towel>', '[CLOSE] <bathroom_cabinet>', '[WALK] <mirror>', '[WIPE] <mirror>', '[FIND] <trashcan>', '[PUTBACK] <paper_towel> <trashcan>']
}
|
Scene_Objects:['pot', 'toilet', 'jelly', 'band-aids', 'chair', 'drawing', 'cpuscreen', 'bathroom_counter', 'floor', 'shampoo', 'ceiling', 'desk', 'keyboard', 'purse', 'brush', 'toaster', 'shoe-shine_kit', 'plate', 'faucet', 'mousepad', 'towel', 'dining_room', 'wall', 'photoframe', 'newspaper', 'coffe_maker', 'toothbrush_holder', 'mouse', 'door', 'home_office']
Task:Go to toilet
response:
| {
"thoughts": "I walk into the bathroom. I walk over to the toilet. I sit on the toilet.",
"actions": [
"[WALK] <bathroom>",
"[WALK] <toilet>",
"[SIT] <toilet>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'freezer', 'ceilingfan', 'drawing', 'knifeblock', 'floor', 'ceiling', 'phone', 'tea_bag', 'computer', 'button', 'bookmark', 'ceilinglamp', 'tvstand', 'balanceball', 'window', 'foundation', 'couch', 'clothes_hat', 'wall', 'oven', 'bench', 'television', 'curtain', 'vacuum_cleaner', 'dresser', 'maindoor', 'spoon', 'home_office', 'table']
Task:Watch TV
response:
{
"thoughts": walk to living room, find the television, switch on the television, find the sofa, walk to sofa, sit in sofa, watch the television,
"actions": ['[WALK] <home_office>', '[WALK] <television>', '[FIND] <television>', '[SWITCHON] <television>', '[FIND] <couch>', '[WALK] <couch>', '[SIT] <couch>', '[TURNTO] <television>', '[WATCH] <television>']
}
|
Scene_Objects:['pot', 'tablelamp', 'chair', 'bathroom', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'ceilinglamp', 'deck_of_cards', 'balanceball', 'mousepad', 'dustpan', 'wall', 'nightstand', 'photoframe', 'wine', 'doorjamb', 'mouse', 'dresser', 'light', 'colander']
Task:Browse internet
response:
| {
"thoughts": "Go find a computer, turn it on, click google, and browse the internet.",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>",
"[TURNTO] <computer>",
"[LOOKAT] <computer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'tablelamp', 'chair', 'freezer', 'drawing', 'orchid', 'floor', 'ceiling', 'desk', 'bedroom', 'bookshelf', 'computer', 'mat', 'shoe-shine_kit', 'oil', 'pillow', 'wall', 'pasta', 'powersocket', 'cup', 'food_food', 'video_game_controller', 'toothbrush_holder', 'television', 'doorjamb', 'kettle', 'door', 'light', 'clothes_skirt', 'coffee']
Task:Turn on light
response:
{
"thoughts": switching on light in an area,
"actions": ['[WALK] <home_office>', '[WALK] <light>', '[FIND] <light>', '[SWITCHON] <light>']
}
|
Scene_Objects:['tablelamp', 'bathroom', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'remote_control', 'purse', 'button', 'ceilinglamp', 'tvstand', 'balanceball', 'window', 'homework', 'couch', 'dining_room', 'wall', 'bench', 'food_food', 'coffe_maker', 'television', 'toilet_paper', 'wooden_spoon', 'curtain', 'pencil', 'shoes', 'table', 'light', 'vase']
Task:Change TV channel
response:
| {
"thoughts": "I go the living room, look for my tv remote control, take the control and then sit down on the couch and turn on the TV and push the buttons on the remote control to change the channel.",
"actions": [
"[WALK] <home_office>",
"[WALK] <remote_control>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[FIND] <couch>",
"[SIT] <couch>",
"[FIND] <television>",
"[SWITCHON] <television>",
"[FIND] <button>",
"[PUSH] <button>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_pizza', 'drawing', 'orchid', 'floor', 'ceiling', 'phone', 'computer', 'tvstand', 'plate', 'faucet', 'shower', 'couch', 'pillow', 'dining_room', 'wall', 'pasta', 'newspaper', 'novel', 'check', 'toilet_paper', 'food_food', 'doorjamb', 'towel_rack', 'kitchen_counter', 'cupboard', 'controller', 'door', 'table', 'light', 'dishwasher']
Task:Relax on sofa
response:
{
"thoughts": I walk into the living room. I cross the room to the couch. I sit down and relax.,
"actions": ['[WALK] <home_office>', '[WALK] <couch>', '[SIT] <couch>']
}
|
Scene_Objects:['pot', 'walllamp', 'knifeblock', 'orchid', 'cpuscreen', 'drawing', 'bathroom_counter', 'floor', 'ceiling', 'shampoo', 'food_bacon', 'table_cloth', 'stovefan', 'faucet', 'dining_room', 'wall', 'oven', 'photoframe', 'novel', 'coffe_maker', 'drinking_glass', 'wine', 'shaving_cream', 'kitchen_counter', 'cupboard', 'maindoor', 'sink', 'bathroom_cabinet', 'dishwasher', 'scrabble']
Task:Drink
response:
| {
"thoughts": "First I go into the kitchen, open the cupboard and find a glass.Then I close the cupboard. Then I go to the sink, turn the faucet on fill the glass under the faucet. I turn off the faucet and then I drink.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <cupboard>",
"[FIND] <cupboard>",
"[OPEN] <cupboard>",
"[FIND] <drinking_glass>",
"[GRAB] <drinking_glass>",
"[WALK] <sink>",
"[FIND] <faucet>",
"[SWITCHON] <faucet>",
"[PUTBACK] <drinking_glass> <sink>",
"[GRAB] <drinking_glass>",
"[DRINK] <drinking_glass>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'juice', 'toilet', 'chair', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'toaster', 'ceilinglamp', 'stovefan', 'balanceball', 'mat', 'mousepad', 'oil', 'wall', 'nightstand', 'photoframe', 'doorjamb', 'mouse', 'light', 'dishwasher']
Task:Work
response:
{
"thoughts": Go to office, sit at desk, turn on computer, enter password, open application and begin work,
"actions": ['[WALK] <home_office>', '[WALK] <computer>', '[FIND] <computer>', '[SWITCHON] <computer>', '[FIND] <keyboard>', '[TYPE] <keyboard>', '[FIND] <mouse>', '[PUSH] <mouse>', '[TURNTO] <computer>', '[LOOKAT] <computer>']
}
|
Scene_Objects:['spectacles', 'food_bread', 'filing_cabinet', 'tablelamp', 'chair', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'ceilinglamp', 'deck_of_cards', 'balanceball', 'mousepad', 'food_noodles', 'pillow', 'wall', 'nightstand', 'oven', 'basket_for_clothes', 'doorjamb', 'towel_rack', 'mouse', 'light']
Task:Work
response:
| {
"thoughts": "walk to home office ,find computer ,switch on computer and work",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_salt', 'microwave', 'trashcan', 'drawing', 'floor', 'dvd_player', 'desk', 'envelope', 'ceilinglamp', 'tvstand', 'mat', 'window', 'mousepad', 'couch', 'pillow', 'dining_room', 'broom', 'wall', 'nightstand', 'cup', 'check', 'television', 'coin', 'colander', 'after_shave', 'cd', 'cupboard', 'table', 'vase', 'soap']
Task:Relax on sofa
response:
{
"thoughts": Go to the living room. Find the sofa. Sit on the sofa. Close your eyes. Breath calmly. Relax,
"actions": ['[WALK] <home_office>', '[WALK] <couch>', '[FIND] <couch>', '[SIT] <couch>']
}
|
Scene_Objects:['filing_cabinet', 'juice', 'cat', 'food_butter', 'orchid', 'hanger', 'floor', 'shampoo', 'phone', 'mat', 'shower', 'electric_shaver', 'dustpan', 'microphone', 'towel', 'pillow', 'dining_room', 'wall', 'bench', 'tape', 'food_cheese', 'check', 'wine', 'curtain', 'vacuum_cleaner', 'dresser', 'sink', 'napkin', 'bathroom_cabinet', 'vase']
Task:Read book
response:
| {
"thoughts": "Open coffee maker. Put filter in main basket, put ground coffee in filter, fill reservoir with water, turn on coffee maker.",
"actions": [
"[WALK] <home_office>",
"[WALK] <address_book>",
"[FIND] <address_book>",
"[GRAB] <address_book>",
"[READ] <address_book>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['drawing', 'bills', 'bathroom_counter', 'floor', 'ceiling', 'character', 'computer', 'toaster', 'ceilinglamp', 'window', 'food_noodles', 'couch', 'microphone', 'pillow', 'wall', 'newspaper', 'novel', 'coffe_maker', 'television', 'coin', 'toothbrush_holder', 'shoes', 'dresser', 'cupboard', 'laundry_detergent', 'bathroom_cabinet', 'table', 'vase', 'laser_pointer', 'creditcard']
Task:Read book
response:
{
"thoughts": Sit on couch. Grab book. Open book. Read book.,
"actions": ['[WALK] <couch>', '[SIT] <couch>', '[FIND] <novel>', '[GRAB] <novel>', '[READ] <novel>']
}
|
Scene_Objects:['sheets', 'filing_cabinet', 'chair', 'freezer', 'ground_coffee', 'ceilingfan', 'drawing', 'needle', 'floor', 'ceiling', 'bookshelf', 'remote_control', 'computer', 'food_bacon', 'ceilinglamp', 'tvstand', 'balanceball', 'window', 'couch', 'clothes_pants', 'wall', 'toy', 'food_food', 'television', 'oven_mitts', 'curtain', 'diary', 'headset', 'home_office', 'table']
Task:Change TV channel
response:
| {
"thoughts": "I go in the living room. Then I find the remote, sit on the couch and turn on the television. Then I use the remote the change the TV channel.",
"actions": [
"[WALK] <home_office>",
"[WALK] <remote_control>",
"[FIND] <remote_control>",
"[FIND] <couch>",
"[SIT] <couch>",
"[FIND] <television>",
"[SWITCHON] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['glue', 'pot', 'filing_cabinet', 'microwave', 'chair', 'cpuscreen', 'hanger', 'bathroom_counter', 'floor', 'ceiling', 'face', 'toaster', 'stovefan', 'faucet', 'oil', 'towel', 'wall', 'facial_cleanser', 'check', 'coffe_maker', 'tray', 'rag', 'mouse', 'kitchen_counter', 'dresser', 'door', 'sink', 'bathroom_cabinet', 'table', 'soap']
Task:Wash face
response:
{
"thoughts": Turn the faucet on. Wet the face. Take facial cleanser out of packaging. Lather on face. Rinse face with water. Turn off the faucet. Dry face with towel.,
"actions": ['[WALK] <bathroom>', '[WALK] <sink>', '[FIND] <faucet>', '[SWITCHON] <faucet>', '[FIND] <face>', '[WASH] <face>', '[FIND] <rag>', '[GRAB] <rag>', '[RINSE] <rag>', '[FIND] <facial_cleanser>', '[GRAB] <facial_cleanser>', '[POUR] <facial_cleanser> <rag>', '[PUTOBJBACK] <facial_cleanser>', '[SCRUB] <face>', '[RINSE] <rag>', '[PUTOBJBACK] <rag>', '[RINSE] <face>', '[SWITCHOFF] <faucet>', '[FIND] <towel>', '[GRAB] <towel>', '[WIPE] <face>', '[PUTOBJBACK] <towel>']
}
|
Scene_Objects:['chair', 'freezer', 'detergent', 'food_butter', 'trashcan', 'hanger', 'floor', 'phone', 'ceiling', 'bedroom', 'computer', 'button', 'mat', 'laptop', 'plate', 'window', 'wall', 'nightstand', 'bed', 'check', 'curtain', 'doorjamb', 'dresser', 'cupboard', 'controller', 'door', 'light', 'dishwasher', 'laser_pointer', 'shoe_rack']
Task:Pick up phone
response:
| {
"thoughts": "first i will go to home office and switch on the lights and answer the call after then i will put back it and switch off the lights and close the door",
"actions": [
"[RUN] <home_office>",
"[WALK] <light>",
"[FIND] <light>",
"[SWITCHON] <light>",
"[FIND] <phone>",
"[GRAB] <phone>",
"[PUTOBJBACK] <phone>",
"[SWITCHOFF] <light>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sponge', 'toilet', 'chair', 'freezer', 'band-aids', 'walllamp', 'cat', 'drawing', 'orchid', 'floor', 'ceiling', 'phone', 'bedroom', 'ceilinglamp', 'tvstand', 'clothes_socks', 'couch', 'pillow', 'dining_room', 'wall', 'bench', 'clothes_gloves', 'novel', 'tray', 'doorjamb', 'kitchen_counter', 'door', 'table', 'light', 'dishwasher']
Task:Pet cat
response:
{
"thoughts": walk to entrance hall, look at the sofa, sit on the sofa, look at the cat, allow cat to near with you, touch the cat smoothly on it head, now pet the cat smoothly,
"actions": ['[WALK] <home_office>', '[WALK] <couch>', '[FIND] <couch>', '[TURNTO] <couch>', '[LOOKAT] <couch>', '[SIT] <couch>', '[FIND] <cat>', '[TURNTO] <cat>', '[LOOKAT] <cat>', '[TOUCH] <cat>']
}
|
Scene_Objects:['food_bread', 'scissors', 'mop', 'chair', 'ceilingfan', 'milk', 'toothbrush', 'knifeblock', 'phone', 'bookshelf', 'stovefan', 'toaster', 'tvstand', 'shower', 'couch', 'food_peanut_butter', 'dustpan', 'oven', 'photoframe', 'clothes_gloves', 'bed', 'video_game_controller', 'tray', 'towel_rack', 'clothes_scarf', 'table', 'light', 'soap', 'tooth_paste', 'scrabble']
Task:Wash teeth
response:
| {
"thoughts": "before going to bed or when wake up earily morning brush teeth",
"actions": [
"[WALK] <bathroom>",
"[WALK] <tooth_paste>",
"[FIND] <tooth_paste>",
"[GRAB] <tooth_paste>",
"[FIND] <toothbrush>",
"[GRAB] <toothbrush>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'freezer', 'bathroom', 'food_carrot', 'drawing', 'orchid', 'wallshelf', 'food_egg', 'bathroom_counter', 'shampoo', 'phone', 'mat', 'ceilinglamp', 'shower', 'electric_shaver', 'clothes_socks', 'couch', 'microphone', 'dining_room', 'food_kiwi', 'nightstand', 'wall', 'check', 'food_food', 'after_shave', 'mouse', 'kitchen_counter', 'cupboard', 'home_office', 'laser_pointer']
Task:Turn on light
response:
{
"thoughts": I go home and it's dark so I can not see anything,
"actions": ['[WALK] <home_office>', '[WALK] <floor_lamp>', '[FIND] <floor_lamp>', '[SWITCHON] <floor_lamp>']
}
|
Scene_Objects:['tablelamp', 'measuring_cup', 'toilet', 'drawing', 'bathroom_counter', 'floor', 'ceiling', 'longboard', 'button', 'ceilinglamp', 'faucet', 'oil', 'teeth', 'clothes_hat', 'towel', 'wall', 'photoframe', 'bed', 'doorjamb', 'diary', 'kitchen_counter', 'dishrack', 'coffee', 'hairbrush', 'maindoor', 'sink', 'bathroom_cabinet', 'vase', 'soap', 'hands_both']
Task:Wash hands
response:
| {
"thoughts": "Go to the bathroom. Turn on the sink. Rinse hands. Apply soap and scrub your hands together. Rinse hands again. Turn off faucet. Dry hands on a towel.",
"actions": [
"[WALK] <bathroom>",
"[WALK] <faucet>",
"[FIND] <faucet>",
"[SWITCHON] <faucet>",
"[FIND] <hands_both>",
"[RINSE] <hands_both>",
"[FIND] <soap>",
"[GRAB] <soap>",
"[SCRUB] <hands_both>",
"[PUTOBJBACK] <soap>",
"[RINSE] <hands_both>",
"[SWITCHOFF] <faucet>",
"[FIND] <towel>",
"[GRAB] <towel>",
"[WIPE] <hands_both>",
"[PUTOBJBACK] <towel>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['fork', 'filing_cabinet', 'blender', 'chair', 'freezer', 'walllamp', 'nail_polish', 'detergent', 'drawing', 'knife', 'wallshelf', 'bathroom_counter', 'floor', 'desk', 'computer', 'mat', 'ceilinglamp', 'plate', 'faucet', 'mousepad', 'candle', 'microphone', 'bowl', 'wall', 'nightstand', 'bench', 'bed', 'keys', 'food_food', 'table']
Task:Eat
response:
{
"thoughts": I get my prepared dinner set on the dining table, sit down on chair, pick up utensils, eat food,
"actions": ['[WALK] <dining_room>', '[WALK] <food_food>', '[FIND] <food_food>', '[GRAB] <food_food>', '[WALK] <dining_room>', '[WALK] <table>', '[FIND] <table>', '[PUTBACK] <food_food> <table>', '[FIND] <chair>', '[SIT] <chair>', '[FIND] <fork>', '[GRAB] <fork>', '[FIND] <knife>', '[GRAB] <knife>', '[FIND] <plate>', '[EAT] <plate>']
}
|
Scene_Objects:['blender', 'dog', 'band-aids', 'drawing', 'bathroom_counter', 'floor', 'longboard', 'bedroom', 'food_cake', 'computer', 'table_cloth', 'ceilinglamp', 'tvstand', 'faucet', 'mat', 'window', 'couch', 'dustpan', 'clothes_pants', 'bowl', 'wall', 'television', 'oven_mitts', 'after_shave', 'doorjamb', 'towel_rack', 'cards', 'shoes', 'home_office', 'table']
Task:Relax on sofa
response:
| {
"thoughts": "I walk into the living room. I bend down. I sit on the sofa.",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[SIT] <couch>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'chair', 'walllamp', 'bathroom', 'orchid', 'cpuscreen', 'wallshelf', 'hanger', 'clothes_underwear', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'computer', 'ceilinglamp', 'tvstand', 'mat', 'mousepad', 'couch', 'pillow', 'wall', 'nightstand', 'check', 'television', 'doorjamb', 'mouse', 'dresser', 'door', 'light']
Task:Research
response:
{
"thoughts": I read different articles online to gain a better understanding of a particular business, industry, method, terminology, etc.,
"actions": ['[WALK] <home_office>', '[WALK] <computer>', '[FIND] <computer>', '[SWITCHON] <computer>', '[TURNTO] <computer>', '[LOOKAT] <computer>', '[FIND] <check>', '[GRAB] <check>', '[READ] <check>']
}
|
Scene_Objects:['tablelamp', 'chair', 'freezer', 'nail_polish', 'trashcan', 'bills', 'drawing', 'hanger', 'bathroom_counter', 'character', 'bedroom', 'keyboard', 'computer', 'stereo', 'shower', 'form', 'dining_room', 'powersocket', 'oven', 'bench', 'cup', 'coffe_maker', 'mouthwash', 'television', 'coin', 'wooden_spoon', 'tray', 'doorjamb', 'mouse', 'bathroom_cabinet']
Task:Listen to music
response:
| {
"thoughts": "I reach out arm. I put hand on radio. I turn on radio. I pick a station.",
"actions": [
"[WALK] <bedroom>",
"[WALK] <stereo>",
"[TOUCH] <stereo>",
"[TURNTO] <stereo>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'freezer', 'bathroom', 'drawing', 'knifeblock', 'cpuscreen', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'bedroom', 'desk', 'bookshelf', 'shower', 'homework', 'wall', 'powersocket', 'box', 'bed', 'coffe_maker', 'television', 'after_shave', 'doorjamb', 'towel_rack', 'mouse', 'cards', 'cupboard', 'home_office', 'light', 'dishwasher']
Task:Turn on light
response:
{
"thoughts": Walk into living room. Find wall. Find light button. Switch on light button.,
"actions": ['[WALK] <home_office>', '[WALK] <light>', '[FIND] <light>', '[SWITCHON] <light>']
}
|
Scene_Objects:['pot', 'food_bread', 'food_ice_cream', 'toilet', 'chair', 'freezer', 'ceilingfan', 'drawing', 'longboard', 'floor', 'ceiling', 'remote_control', 'ceilinglamp', 'tvstand', 'balanceball', 'mat', 'window', 'oil', 'couch', 'wall', 'bench', 'mouthwash', 'television', 'curtain', 'shaving_cream', 'mouse', 'headset', 'bathroom_cabinet', 'table', 'light']
Task:Watch TV
response:
| {
"thoughts": "go to the living room. find the remote. sit on the couch. turn on the tv.",
"actions": [
"[WALK] <home_office>",
"[WALK] <remote_control>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[WALK] <couch>",
"[SIT] <couch>",
"[FIND] <television>",
"[SWITCHON] <television>",
"[TURNTO] <television>",
"[WATCH] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sheets', 'juice', 'microwave', 'freezer', 'bills', 'floor', 'ceiling', 'desk', 'tea_bag', 'computer', 'ceilinglamp', 'plate', 'window', 'mousepad', 'couch', 'dining_room', 'broom', 'wall', 'powersocket', 'novel', 'food_food', 'mouthwash', 'television', 'curtain', 'rag', 'towel_rack', 'kitchen_counter', 'comb', 'door', 'laser_pointer']
Task:Put groceries in Fridge
response:
{
"thoughts": Walk to the kitchen. Open the fridge, take the groceries and put in it. Closed the fridge,,
"actions": ['[WALK] <dining_room>', '[WALK] <freezer>', '[FIND] <freezer>', '[OPEN] <freezer>', '[FIND] <food_food>', '[GRAB] <food_food>', '[PUTIN] <food_food> <freezer>', '[CLOSE] <freezer>']
}
|
Scene_Objects:['food_pizza', 'filing_cabinet', 'tablelamp', 'chair', 'freezer', 'detergent', 'hanger', 'floor', 'ceiling', 'phone', 'purse', 'computer', 'mail', 'toaster', 'tvstand', 'mousepad', 'wall', 'check', 'ice', 'tray', 'doorjamb', 'towel_rack', 'coffee_pot', 'cards', 'dresser', 'cupboard', 'door', 'napkin', 'light', 'scrabble']
Task:Pick up phone
response:
| {
"thoughts": "I walk over to the phone. .I reach my arm out. I open my hand. I place my hand on the phone. I close my fingers around the phone. I pick up the phone.",
"actions": [
"[WALK] <phone>",
"[TOUCH] <phone>",
"[GRAB] <phone>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'tablelamp', 'chair', 'ground_coffee', 'wallshelf', 'hanger', 'floor', 'stamp', 'desk', 'bookshelf', 'tea_bag', 'envelope', 'computer', 'mop_bucket', 'mat', 'shower', 'foundation', 'form', 'wall', 'powersocket', 'cutting_board', 'photoframe', 'novel', 'check', 'television', 'wooden_spoon', 'after_shave', 'doorjamb', 'door', 'light']
Task:Read to child
response:
{
"thoughts": Read a book out loud to a child,
"actions": ['[WALK] <bedroom>', '[WALK] <child>', '[FIND] <child>', '[GREET] <child>', '[WALK] <filing_cabinet>', '[OPEN] <filing_cabinet>', '[FIND] <novel>', '[GRAB] <novel>', '[WALK] <chair>', '[PULL] <chair>', '[SIT] <chair>', '[TURNTO] <novel>', '[LOOKAT] <novel>', '[TURNTO] <novel>', '[POINTAT] <novel>', '[TURNTO] <novel>', '[POINTAT] <novel>', '[READ] <novel>', '[READ] <novel>', '[READ] <novel>']
}
|
Scene_Objects:['pot', 'food_bread', 'freezer', 'food_butter', 'toothbrush', 'knifeblock', 'food_egg', 'cpuscreen', 'longboard', 'bathroom_counter', 'phone', 'ceiling', 'tea_bag', 'plate', 'shower', 'deck_of_cards', 'window', 'electric_shaver', 'dish_soap', 'broom', 'bench', 'powersocket', 'tape', 'mouthwash', 'centerpiece', 'shaving_cream', 'mouse', 'bathroom_cabinet', 'dishwasher', 'coffee']
Task:Wash dishes by hand
response:
| {
"thoughts": "Pick up dish. Put in soapy water. Wash dish.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <dish_soap>",
"[FIND] <dish_soap>",
"[GRAB] <dish_soap>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[PUTBACK] <dish_soap> <plate>",
"[SCRUB] <plate>",
"[RINSE] <plate>",
"[WIPE] <plate>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sponge', 'pot', 'blender', 'beer', 'microwave', 'knifeblock', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'table_cloth', 'toaster', 'plate', 'faucet', 'tvstand', 'window', 'dish_soap', 'food_orange', 'wall', 'cup', 'bed', 'food_food', 'tray', 'wooden_spoon', 'mouse', 'pencil', 'kitchen_counter', 'dishrack', 'cupboard', 'sink']
Task:Wash dirty dishes
response:
{
"thoughts": Wet the sponge, apply soap to sponge, scrub dishes, rinse dishes, place on dish rack.,
"actions": ['[WALK] <dining_room>', '[WALK] <sink>', '[FIND] <sponge>', '[GRAB] <sponge>', '[FIND] <faucet>', '[SWITCHON] <faucet>', '[PUTBACK] <sponge> <sink>', '[GRAB] <sponge>', '[FIND] <dish_soap>', '[GRAB] <dish_soap>', '[POUR] <dish_soap> <sponge>', '[PUTOBJBACK] <dish_soap>', '[FIND] <plate>', '[GRAB] <plate>', '[SCRUB] <plate>', '[RINSE] <plate>', '[FIND] <dishrack>', '[PUTBACK] <plate> <dishrack>', '[FIND] <plate>', '[GRAB] <plate>', '[SCRUB] <plate>', '[RINSE] <plate>', '[PUTBACK] <plate> <dishrack>', '[RINSE] <sponge>', '[PUTOBJBACK] <sponge>', '[SWITCHOFF] <faucet>']
}
|
Scene_Objects:['tea', 'freezer', 'walllamp', 'knifeblock', 'kitchen_cabinet', 'floor', 'ceiling', 'dvd_player', 'desk', 'table_cloth', 'ceilinglamp', 'plate', 'mousepad', 'couch', 'candle', 'tape', 'light_bulb', 'novel', 'coin', 'water', 'drinking_glass', 'wine', 'tray', 'coffee_pot', 'mouse', 'cd', 'conditioner', 'maindoor', 'napkin', 'home_office']
Task:Drink
response:
| {
"thoughts": "You walk to the kitchen and open the kitchen cabinet and grab a drinking glass. You find the water and pour the water into the drinking glass. You drink the water from the glass.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <kitchen_cabinet>",
"[FIND] <kitchen_cabinet>",
"[OPEN] <kitchen_cabinet>",
"[FIND] <drinking_glass>",
"[GRAB] <drinking_glass>",
"[WALK] <water>",
"[FIND] <water>",
"[GRAB] <water>",
"[POUR] <water> <drinking_glass>",
"[DRINK] <drinking_glass>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'tablelamp', 'dog', 'freezer', 'dough', 'knifeblock', 'knife', 'hanger', 'bathroom_counter', 'ceiling', 'toaster', 'tvstand', 'faucet', 'couch', 'pasta', 'powersocket', 'oven', 'photoframe', 'novel', 'television', 'wooden_spoon', 'curtain', 'coffee_pot', 'mouse', 'dresser', 'door', 'sink', 'home_office', 'light', 'bathroom_cabinet']
Task:Greet guests
response:
{
"thoughts": Say hi,
"actions": ['[WALK] <home_office>', '[WALK] <child>', '[FIND] <child>', '[GREET] <child>']
}
|
Scene_Objects:['sheets', 'scissors', 'mop', 'toilet', 'blow_dryer', 'detergent', 'drawing', 'cpuscreen', 'bathroom_counter', 'floor', 'ceiling', 'desk', 'keyboard', 'toaster', 'tvstand', 'shower', 'mousepad', 'pillow', 'wall', 'powersocket', 'oven', 'cup', 'photoframe', 'bed', 'food_onion', 'doorjamb', 'shaving_cream', 'mouse', 'kitchen_counter', 'bathroom_cabinet']
Task:Go to toilet
response:
| {
"thoughts": "I walk to bathroom, I open the door of the bathroom, and sit on the toilet.",
"actions": [
"[WALK] <bathroom>",
"[WALK] <toilet>",
"[FIND] <toilet>",
"[SIT] <toilet>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'crayon', 'tea', 'mop', 'microwave', 'toilet', 'toothbrush', 'drawing', 'knifeblock', 'orchid', 'phone', 'ceiling', 'computer', 'stovefan', 'toaster', 'tvstand', 'window', 'floor_lamp', 'couch', 'clothes_hat', 'photoframe', 'after_shave', 'food_onion', 'headset', 'chessboard', 'cupboard', 'controller', 'door', 'bathroom_cabinet', 'clothes_skirt']
Task:Switch on lamp
response:
{
"thoughts": Press the button so the light comes on.,
"actions": ['[WALK] <bedroom>', '[WALK] <floor_lamp>', '[FIND] <floor_lamp>', '[SWITCHON] <floor_lamp>']
}
|
Scene_Objects:['sheets', 'pot', 'chair', 'freezer', 'food_carrot', 'cpuscreen', 'bathroom_counter', 'floor', 'ceiling', 'thread', 'mail', 'ceilinglamp', 'mat', 'iron', 'balanceball', 'faucet', 'food_orange', 'pillow', 'wall', 'pasta', 'box', 'cup', 'newspaper', 'food_food', 'check', 'toilet_paper', 'chessboard', 'maindoor', 'napkin', 'creditcard']
Task:Put groceries in Fridge
response:
| {
"thoughts": "Putting groceries in refridgerator",
"actions": [
"[WALK] <dining_room>",
"[WALK] <food_food>",
"[FIND] <food_food>",
"[GRAB] <food_food>",
"[FIND] <freezer>",
"[OPEN] <freezer>",
"[PUTIN] <food_food> <freezer>",
"[CLOSE] <freezer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_pizza', 'toilet', 'freezer', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'envelope', 'table_cloth', 'stovefan', 'ceilinglamp', 'mat', 'shower', 'stereo', 'mousepad', 'towel', 'wall', 'powersocket', 'photoframe', 'bed', 'toilet_paper', 'tray', 'curtain', 'doorjamb', 'dresser', 'controller', 'door', 'food_steak', 'light', 'tooth_paste']
Task:Take shower
response:
{
"thoughts": walk to bathroom , find shower , take a shower,
"actions": ['[WALK] <bathroom>', '[WALK] <shower>', '[FIND] <shower>']
}
|
Scene_Objects:['filing_cabinet', 'food_salt', 'chair', 'blow_dryer', 'bathroom', 'drawing', 'knifeblock', 'bathroom_counter', 'floor', 'shampoo', 'bookshelf', 'purse', 'keyboard', 'ceilinglamp', 'deck_of_cards', 'shower', 'window', 'couch', 'wall', 'powersocket', 'tape', 'photoframe', 'television', 'food_rice', 'after_shave', 'controller', 'conditioner', 'door', 'table', 'scrabble']
Task:Relax on sofa
response:
| {
"thoughts": "I walk into the living room and I lie down on the sofa.",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[LIE] <couch>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'food_pizza', 'tablelamp', 'food_ice_cream', 'crayon', 'shampoo', 'keyboard', 'mail', 'bookmark', 'stovefan', 'faucet', 'pillow', 'dining_room', 'broom', 'wall', 'love_seat', 'photoframe', 'novel', 'food_food', 'wine', 'tray', 'after_shave', 'shaving_cream', 'kettle', 'headset', 'pajamas', 'laundry_detergent', 'door', 'bathroom_cabinet', 'shoe_rack']
Task:Read book
response:
{
"thoughts": Obtain a book from the book shelf. Sit in the easy chair. Read the book. Mark the stopping point with a bookmark. Put the book back on the book shelf.,
"actions": ['[WALK] <home_office>', '[WALK] <novel>', '[FIND] <novel>', '[GRAB] <novel>', '[FIND] <love_seat>', '[SIT] <love_seat>', '[READ] <novel>', '[FIND] <bookmark>', '[GRAB] <bookmark>', '[PUTBACK] <bookmark> <novel>', '[PUTOBJBACK] <novel>']
}
|
Scene_Objects:['sheets', 'filing_cabinet', 'beer', 'jelly', 'bathroom', 'ceilingfan', 'drawing', 'floor', 'ceiling', 'keyboard', 'clothes_jacket', 'table_cloth', 'toaster', 'ceilinglamp', 'tvstand', 'balanceball', 'window', 'homework', 'food_noodles', 'couch', 'wall', 'oven', 'basket_for_clothes', 'cup', 'television', 'curtain', 'conditioner', 'maindoor', 'table', 'vase']
Task:Watch TV
response:
| {
"thoughts": "Walk into the living room. Sit on the couch. Turn on TV.",
"actions": [
"[WALK] <home_office>",
"[WALK] <television>",
"[FIND] <television>",
"[SWITCHON] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_ice_cream', 'measuring_cup', 'toilet', 'chair', 'drawing', 'food_egg', 'clothes_underwear', 'hanger', 'bathroom_counter', 'floor', 'ceiling', 'stamp', 'bookshelf', 'button', 'shoe-shine_kit', 'shower', 'window', 'broom', 'wall', 'wall_clock', 'powersocket', 'bench', 'standingmirror', 'novel', 'coin', 'doorjamb', 'door', 'light', 'table', 'creditcard']
Task:Read
response:
{
"thoughts": We pick out a book from the bookshelf and read it.,
"actions": ['[WALK] <novel>', '[WALK] <bookshelf>', '[FIND] <novel>', '[GRAB] <novel>', '[FIND] <chair>', '[SIT] <chair>', '[READ] <novel>']
}
|
Scene_Objects:['pot', 'food_pizza', 'tablelamp', 'food_salt', 'freezer', 'walllamp', 'trashcan', 'food_carrot', 'knifeblock', 'floor', 'ceiling', 'food_bacon', 'stovefan', 'ceilinglamp', 'toaster', 'mousepad', 'wall', 'nightstand', 'oven', 'toy', 'box', 'photoframe', 'keys', 'toothbrush_holder', 'tray', 'kitchen_counter', 'hair', 'door', 'sink', 'dishwasher']
Task:Cook some food
response:
| {
"thoughts": "go to the kitchen and cook some thing to eat",
"actions": [
"[WALK] <dining_room>",
"[WALK] <freezer>",
"[FIND] <freezer>",
"[OPEN] <freezer>",
"[FIND] <oven>",
"[OPEN] <oven>",
"[FIND] <food_pizza>",
"[GRAB] <food_pizza>",
"[PUTIN] <food_pizza> <oven>",
"[SWITCHON] <oven>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['band-aids', 'trashcan', 'wallshelf', 'clothes_underwear', 'desk', 'keyboard', 'mail', 'computer', 'tvstand', 'shower', 'window', 'clothes_socks', 'couch', 'food_peanut_butter', 'pillow', 'basket_for_clothes', 'keys', 'coffe_maker', 'television', 'food_donut', 'tray', 'rag', 'headset', 'clothes_scarf', 'bathtub', 'sink', 'light', 'dishwasher', 'soap', 'washing_machine']
Task:Wash clothes
response:
{
"thoughts": Get dirty clothes out of basket for clothes in bedroom, carry dirty clothes to laundry room, place dirty clothes in washing machine, add laundry detergent to washing machine, turn washing machine on, set type of wash, start washing machine,
"actions": ['[WALK] <bedroom>', '[WALK] <basket_for_clothes>', '[FIND] <basket_for_clothes>', '[OPEN] <basket_for_clothes>', '[GRAB] <basket_for_clothes>', '[WALK] <bathroom>', '[WALK] <washing_machine>', '[FIND] <washing_machine>', '[OPEN] <washing_machine>', '[PUTBACK] <basket_for_clothes> <washing_machine>', '[FIND] <soap>', '[GRAB] <soap>', '[PUTBACK] <soap> <washing_machine>', '[FIND] <washing_machine>', '[CLOSE] <washing_machine>', '[SWITCHON] <washing_machine>']
}
|
Scene_Objects:['tablelamp', 'microwave', 'freezer', 'bathroom', 'cpuscreen', 'desk', 'tea_bag', 'alcohol', 'tvstand', 'oven', 'tape', 'cutting_board', 'cup', 'clothes_gloves', 'novel', 'toilet_paper', 'television', 'coffee_filter', 'wooden_spoon', 'mouse', 'cd', 'controller', 'maindoor', 'sink', 'home_office', 'light', 'dishwasher', 'soap', 'dirt', 'shoe_rack']
Task:Listen to music
response:
| {
"thoughts": "Walk into living room. Turn on radio.",
"actions": [
"[WALK] <home_office>",
"[WALK] <stereo>",
"[FIND] <stereo>",
"[SWITCHON] <stereo>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['fork', 'pot', 'toilet', 'chair', 'freezer', 'food_butter', 'food_carrot', 'drawing', 'floor', 'ceiling', 'keyboard', 'mat', 'ceilinglamp', 'shower', 'oil', 'food_peanut_butter', 'food_orange', 'nightstand', 'wall', 'powersocket', 'photoframe', 'food_food', 'food_onion', 'doorjamb', 'pencil', 'dresser', 'door', 'napkin', 'light', 'shoe_rack']
Task:Put groceries in Fridge
response:
{
"thoughts": put away groceries in fridge. open the door, put on shelf, close the door.,
"actions": ['[WALK] <dining_room>', '[WALK] <freezer>', '[FIND] <freezer>', '[OPEN] <freezer>', '[FIND] <food_food>', '[GRAB] <food_food>', '[PUTIN] <food_food> <freezer>', '[CLOSE] <freezer>']
}
|
Scene_Objects:['tablelamp', 'toilet', 'chair', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'ceilinglamp', 'balanceball', 'mousepad', 'couch', 'wall', 'nightstand', 'toy', 'photoframe', 'curtain', 'rag', 'doorjamb', 'mouse', 'dresser', 'door', 'light']
Task:Browse internet
response:
| {
"thoughts": "walk into Living room, look for computer, walk to computer, turn on computer, grab mouse, click mouse",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[TURNTO] <computer>",
"[LOOKAT] <computer>",
"[WALK] <computer>",
"[TURNTO] <computer>",
"[FIND] <mouse>",
"[GRAB] <mouse>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_bread', 'chair', 'detergent', 'food_carrot', 'clothes_underwear', 'hanger', 'floor', 'ceiling', 'phone', 'computer', 'bookmark', 'toaster', 'balanceball', 'wall', 'powersocket', 'photoframe', 'coffee_filter', 'tray', 'doorjamb', 'cards', 'mouse', 'shaving_cream', 'shoes', 'dresser', 'cupboard', 'door', 'napkin', 'home_office', 'light', 'vase']
Task:Turn on light
response:
{
"thoughts": switching on light in an area,
"actions": ['[WALK] <home_office>', '[WALK] <light>', '[FIND] <light>', '[SWITCHON] <light>']
}
|
Scene_Objects:['glue', 'scissors', 'tablelamp', 'freezer', 'trashcan', 'drawing', 'floor', 'shampoo', 'folder', 'bookshelf', 'purse', 'computer', 'window', 'wall', 'powersocket', 'cutting_board', 'cup', 'bed', 'photoframe', 'television', 'curtain', 'food_onion', 'shaving_cream', 'mouse', 'pencil', 'hair', 'maindoor', 'clothes_scarf', 'home_office', 'colander']
Task:Go to sleep
response:
| {
"thoughts": "go t the bedroom and go to sleep",
"actions": [
"[WALK] <bedroom>",
"[WALK] <bed>",
"[FIND] <bed>",
"[LIE] <bed>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['dry_pasta', 'walllamp', 'cat', 'drawing', 'orchid', 'cpuscreen', 'hanger', 'bathroom_counter', 'floor', 'character', 'envelope', 'computer', 'toaster', 'mat', 'tvstand', 'ceilinglamp', 'shower', 'window', 'mousepad', 'couch', 'pillow', 'wall', 'wall_clock', 'oven', 'coffe_maker', 'television', 'curtain', 'hair', 'bathroom_cabinet', 'table']
Task:Pet cat
response:
{
"thoughts": I would walk into the living room. Then sit on the couch and pet the cat.,
"actions": ['[WALK] <home_office>', '[WALK] <couch>', '[SIT] <couch>', '[FIND] <cat>', '[TOUCH] <cat>']
}
|
Scene_Objects:['pot', 'toilet', 'jelly', 'ceilingfan', 'drawing', 'hanger', 'floor', 'stamp', 'desk', 'ceilinglamp', 'balanceball', 'faucet', 'window', 'couch', 'clothes_pants', 'pillow', 'wall', 'nightstand', 'bed', 'photoframe', 'television', 'oven_mitts', 'food_rice', 'doorjamb', 'cards', 'shaving_cream', 'sink', 'table', 'light', 'shoe_rack']
Task:Relax on sofa
response:
| {
"thoughts": "go to living room and lie down on the sofa.",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[LIE] <couch>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'microwave', 'toilet', 'chair', 'freezer', 'bathroom', 'orchid', 'floor', 'bedroom', 'bookshelf', 'toaster', 'tvstand', 'faucet', 'shower', 'stereo', 'closetdrawer', 'couch', 'broom', 'wall', 'wall_clock', 'photoframe', 'bed', 'television', 'mouse', 'kitchen_counter', 'comb', 'dresser', 'sink', 'spoon', 'clothes_skirt']
Task:Go to toilet
response:
{
"thoughts": I walk into the bath room and go to the toilet,
"actions": ['[WALK] <bathroom>', '[WALK] <toilet>', '[FIND] <toilet>', '[SIT] <toilet>']
}
|
Scene_Objects:['pot', 'chair', 'ground_coffee', 'food_butter', 'toothbrush', 'knifeblock', 'drawing', 'phone', 'keyboard', 'bookshelf', 'table_cloth', 'mat', 'plate', 'electric_shaver', 'window', 'pillow', 'nightstand', 'oven', 'newspaper', 'clothes_gloves', 'photoframe', 'toilet_paper', 'water', 'centerpiece', 'towel_rack', 'kettle', 'pencil', 'clothes_scarf', 'tooth_paste', 'teeth']
Task:Wash teeth
response:
| {
"thoughts": "Walk to bathroom. Find tooth paste, tooth brush and water. Squeeze tooth paste on tooth brush. Wash teeth.",
"actions": [
"[WALK] <bathroom>",
"[WALK] <tooth_paste>",
"[FIND] <tooth_paste>",
"[FIND] <toothbrush>",
"[FIND] <water>",
"[SQUEEZE] <tooth_paste>",
"[FIND] <teeth>",
"[WASH] <teeth>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sponge', 'pot', 'blender', 'beer', 'microwave', 'knifeblock', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'table_cloth', 'toaster', 'plate', 'faucet', 'tvstand', 'window', 'dish_soap', 'food_orange', 'wall', 'cup', 'bed', 'food_food', 'tray', 'wooden_spoon', 'mouse', 'pencil', 'kitchen_counter', 'dishrack', 'cupboard', 'sink']
Task:Wash dirty dishes
response:
{
"thoughts": Wet the sponge, apply soap to sponge, scrub dishes, rinse dishes, place on dish rack.,
"actions": ['[WALK] <dining_room>', '[WALK] <sink>', '[FIND] <sponge>', '[GRAB] <sponge>', '[FIND] <faucet>', '[SWITCHON] <faucet>', '[PUTBACK] <sponge> <sink>', '[GRAB] <sponge>', '[FIND] <dish_soap>', '[GRAB] <dish_soap>', '[POUR] <dish_soap> <sponge>', '[PUTOBJBACK] <dish_soap>', '[FIND] <plate>', '[GRAB] <plate>', '[SCRUB] <plate>', '[RINSE] <plate>', '[FIND] <dishrack>', '[PUTBACK] <plate> <dishrack>', '[FIND] <plate>', '[GRAB] <plate>', '[SCRUB] <plate>', '[RINSE] <plate>', '[PUTBACK] <plate> <dishrack>', '[RINSE] <sponge>', '[PUTOBJBACK] <sponge>', '[SWITCHOFF] <faucet>']
}
|
Scene_Objects:['food_bread', 'tablelamp', 'walllamp', 'ground_coffee', 'ceilingfan', 'drawing', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'desk', 'mat', 'faucet', 'couch', 'pillow', 'wall', 'pasta', 'wall_clock', 'cup', 'photoframe', 'bed', 'television', 'water', 'wine', 'curtain', 'paper_towel', 'sink', 'bathroom_cabinet', 'soap', 'hands_both']
Task:Wash hands
response:
| {
"thoughts": "I walk into the bathroom. I turn on the water. I put soap on my hands. I rinse my hands. I turn off the water. I dry my hands.",
"actions": [
"[WALK] <bathroom>",
"[WALK] <sink>",
"[FIND] <water>",
"[TURNTO] <water>",
"[FIND] <soap>",
"[GRAB] <soap>",
"[FIND] <hands_both>",
"[SCRUB] <hands_both>",
"[RINSE] <hands_both>",
"[TURNTO] <water>",
"[FIND] <paper_towel>",
"[GRAB] <paper_towel>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['scissors', 'tablelamp', 'chair', 'freezer', 'walllamp', 'drawing', 'wallshelf', 'hanger', 'floor', 'ceiling', 'desk', 'keyboard', 'computer', 'mat', 'tvstand', 'ceilinglamp', 'faucet', 'shower', 'closetdrawer', 'couch', 'pillow', 'wall', 'nightstand', 'photoframe', 'newspaper', 'television', 'doorjamb', 'dresser', 'soap', 'instrument_violin']
Task:Watch TV
response:
{
"thoughts": I sit on my couch and watch TV in my living room,
"actions": ['[WALK] <home_office>', '[WALK] <television>', '[SWITCHON] <television>', '[TURNTO] <television>', '[LOOKAT] <television>', '[TOUCH] <television>', '[TURNTO] <television>', '[LOOKAT] <television>', '[TOUCH] <television>', '[TURNTO] <television>', '[LOOKAT] <television>', '[WALK] <couch>', '[SIT] <couch>', '[TURNTO] <television>', '[LOOKAT] <television>', '[WATCH] <television>', '[WATCH] <television>', '[WATCH] <television>']
}
|
Scene_Objects:['spectacles', 'pot', 'tablelamp', 'chair', 'dog', 'walllamp', 'ceilingfan', 'knifeblock', 'needle', 'bathroom_counter', 'floor', 'computer', 'shoe-shine_kit', 'dustpan', 'box', 'cup', 'bed', 'keys', 'television', 'food_donut', 'tray', 'curtain', 'doorjamb', 'kettle', 'kitchen_counter', 'shoes', 'comb', 'sink', 'bathroom_cabinet', 'board_game']
Task:Brush teeth
response:
| {
"thoughts": "go to the bathroom and brush your teeth",
"actions": [
"[WALK] <bathroom>",
"[WALK] <toothbrush>",
"[FIND] <toothbrush>",
"[GRAB] <toothbrush>",
"[WALK] <tooth_paste>",
"[FIND] <tooth_paste>",
"[GRAB] <tooth_paste>",
"[POUR] <tooth_paste> <toothbrush>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'tablelamp', 'chair', 'walllamp', 'knifeblock', 'cpuscreen', 'wallshelf', 'bathroom_counter', 'floor', 'ceiling', 'ceilinglamp', 'faucet', 'couch', 'dustpan', 'pillow', 'towel', 'wall', 'nightstand', 'oven', 'box', 'bed', 'keys', 'check', 'toilet_paper', 'television', 'kitchen_counter', 'dresser', 'instrument_guitar', 'soap', 'tooth_paste']
Task:Relax on sofa
response:
{
"thoughts": Go to living room. Find sofa. Sit on sofa.,
"actions": ['[WALK] <home_office>', '[WALK] <couch>', '[FIND] <couch>', '[SIT] <couch>']
}
|
Scene_Objects:['food_pizza', 'tablelamp', 'jelly', 'ceilingfan', 'knifeblock', 'cpuscreen', 'bathroom_counter', 'floor', 'bookshelf', 'computer', 'clothes_jacket', 'piano_bench', 'stovefan', 'toaster', 'tvstand', 'ceilinglamp', 'balanceball', 'stereo', 'brush', 'food_jam', 'clothes_hat', 'nightstand', 'pasta', 'powersocket', 'bench', 'bed', 'oven_mitts', 'tray', 'towel_rack', 'home_office']
Task:Listen to music
response:
| {
"thoughts": "I walk to the radio. I switch on the radio and listen",
"actions": [
"[WALK] <home_office>",
"[WALK] <stereo>",
"[SWITCHON] <stereo>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['walllamp', 'orchid', 'drawing', 'wallshelf', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'bookshelf', 'remote_control', 'envelope', 'clothes_jacket', 'mat', 'tvstand', 'ceilinglamp', 'shower', 'window', 'couch', 'pillow', 'wall', 'powersocket', 'bench', 'photoframe', 'bed', 'television', 'curtain', 'doorjamb', 'door', 'light', 'table']
Task:Watch TV
response:
{
"thoughts": Enter the living room, find TV remote, grab TV remote, switch on TV, find sofa, sit down on the sofa, watch TV,
"actions": ['[WALK] <home_office>', '[WALK] <remote_control>', '[FIND] <remote_control>', '[GRAB] <remote_control>', '[FIND] <television>', '[SWITCHON] <television>', '[FIND] <couch>', '[WALK] <couch>', '[SIT] <couch>', '[TURNTO] <television>', '[WATCH] <television>']
}
|
Scene_Objects:['sponge', 'chair', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'desk', 'keyboard', 'remote_control', 'food_cake', 'envelope', 'computer', 'ceilinglamp', 'tvstand', 'balanceball', 'faucet', 'window', 'couch', 'wall', 'wall_clock', 'light_bulb', 'television', 'curtain', 'hair', 'food_steak', 'sink', 'bathroom_cabinet', 'table', 'dishwasher']
Task:Watch TV
response:
| {
"thoughts": "I enter the living room and sit on the couch. I grab the remote and turn on the TV.",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[SIT] <couch>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[FIND] <television>",
"[SWITCHON] <television>",
"[TURNTO] <television>",
"[WATCH] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'toilet', 'chair', 'walllamp', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'phone', 'desk', 'bookshelf', 'keyboard', 'envelope', 'computer', 'bookmark', 'ceilinglamp', 'balanceball', 'mousepad', 'wall', 'nightstand', 'photoframe', 'check', 'doorjamb', 'mouse', 'dresser', 'cupboard', 'door', 'light']
Task:Work
response:
{
"thoughts": I go to the home office to work on my project. I turn on the light and walk to my desk. I turn on the computer. Then I look through the documents. After I'm done, I turn off the computer.,
"actions": ['[WALK] <home_office>', '[WALK] <light>', '[FIND] <light>', '[SWITCHON] <light>', '[WALK] <desk>', '[FIND] <computer>', '[SWITCHON] <computer>', '[FIND] <check>', '[TURNTO] <check>', '[LOOKAT] <check>', '[SWITCHOFF] <computer>']
}
|
Scene_Objects:['freezer', 'trashcan', 'food_carrot', 'orchid', 'bathroom_counter', 'floor', 'ceiling', 'desk', 'alcohol', 'button', 'ceilinglamp', 'plate', 'clothes_socks', 'couch', 'food_orange', 'wall', 'bench', 'powersocket', 'photoframe', 'light_bulb', 'food_food', 'coffee_filter', 'drinking_glass', 'water', 'towel_rack', 'dishrack', 'cupboard', 'clothes_dress', 'light', 'colander']
Task:Drink
response:
| {
"thoughts": "im feeling thirsty im going to kitchen and open the freezer where i can find some cold water and im going to drink it to end up my thirst",
"actions": [
"[WALK] <dining_room>",
"[WALK] <freezer>",
"[FIND] <freezer>",
"[OPEN] <freezer>",
"[FIND] <drinking_glass>",
"[GRAB] <drinking_glass>",
"[WALK] <water>",
"[FIND] <water>",
"[GRAB] <water>",
"[POUR] <water> <drinking_glass>",
"[DRINK] <drinking_glass>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['fork', 'filing_cabinet', 'food_bread', 'microwave', 'trashcan', 'drawing', 'knifeblock', 'cpuscreen', 'hanger', 'bathroom_counter', 'bedroom', 'tea_bag', 'tvstand', 'window', 'form', 'pillow', 'broom', 'bench', 'basket_for_clothes', 'newspaper', 'bed', 'check', 'coffe_maker', 'tray', 'door', 'home_office', 'light', 'bathroom_cabinet', 'tooth_paste', 'coffee']
Task:Get dressed
response:
{
"thoughts": I get dressed in my bedroom in the mornings.,
"actions": ['[WALK] <bedroom>', '[WALK] <shoes>', '[FIND] <shoes>', '[GRAB] <shoes>', '[PUTON] <shoes>']
}
|
Scene_Objects:['pot', 'food_bread', 'microwave', 'bathroom', 'ceilingfan', 'drawing', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'toaster', 'mat', 'faucet', 'towel', 'wall', 'bench', 'photoframe', 'light_bulb', 'lighter', 'pencil', 'chessboard', 'dresser', 'hair', 'cupboard', 'sink', 'bathroom_cabinet', 'vase', 'soap', 'hands_both', 'shoe_rack']
Task:Wash hands
response:
| {
"thoughts": "I go to the sink, turn on the water, put soap on my hands, rub them together, then rinse them off and turn off the water. I then towel off my hands.",
"actions": [
"[WALK] <bathroom>",
"[WALK] <sink>",
"[FIND] <faucet>",
"[SWITCHON] <faucet>",
"[FIND] <soap>",
"[GRAB] <soap>",
"[FIND] <hands_both>",
"[WASH] <hands_both>",
"[RINSE] <hands_both>",
"[SWITCHOFF] <faucet>",
"[FIND] <towel>",
"[GRAB] <towel>",
"[FIND] <hands_both>",
"[WIPE] <hands_both>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['jelly', 'trashcan', 'cpuscreen', 'clothes_underwear', 'ceiling', 'folder', 'desk', 'computer', 'stovefan', 'shower', 'mousepad', 'candle', 'clothes_hat', 'dining_room', 'newspaper', 'photoframe', 'food_food', 'video_game_controller', 'oven_mitts', 'curtain', 'pencil', 'kitchen_counter', 'dishrack', 'laundry_detergent', 'door', 'light', 'dishwasher', 'soap', 'tooth_paste', 'washing_machine']
Task:Wash clothes
response:
{
"thoughts": Walk to bathroom. Open washer. Put clothes in washer. Put soap in washer. Close washer. Turn washer on.,
"actions": ['[WALK] <home_office>', '[WALK] <washing_machine>', '[FIND] <washing_machine>', '[OPEN] <washing_machine>', '[FIND] <clothes_underwear>', '[GRAB] <clothes_underwear>', '[PUTBACK] <clothes_underwear> <washing_machine>', '[FIND] <soap>', '[GRAB] <soap>', '[PUTBACK] <soap> <washing_machine>', '[CLOSE] <washing_machine>', '[SWITCHON] <washing_machine>']
}
|
Scene_Objects:['filing_cabinet', 'chair', 'bathroom', 'milk', 'knifeblock', 'cpuscreen', 'floor', 'ceiling', 'phone', 'character', 'desk', 'bookshelf', 'keyboard', 'purse', 'computer', 'shower', 'window', 'electric_shaver', 'mousepad', 'oil', 'couch', 'form', 'wall', 'food_food', 'television', 'rag', 'lighter', 'mouse', 'controller', 'sink']
Task:Pick up phone
response:
| {
"thoughts": "first when i hear the ringing sound i will run to my living room and picks up and i will say hello",
"actions": [
"[RUN] <home_office>",
"[WALK] <chair>",
"[FIND] <chair>",
"[SIT] <chair>",
"[FIND] <phone>",
"[GRAB] <phone>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['dry_pasta', 'food_ice_cream', 'toilet', 'freezer', 'drawing', 'floor', 'ceiling', 'phone', 'stovefan', 'mat', 'ceilinglamp', 'shower', 'bowl', 'nightstand', 'wall', 'powersocket', 'oven', 'photoframe', 'bed', 'food_food', 'coffe_maker', 'wooden_spoon', 'doorjamb', 'pencil', 'comb', 'door', 'light', 'soap', 'scrabble', 'creditcard']
Task:Put groceries in Fridge
response:
{
"thoughts": In the kitchen where I left the bags of groceries, I grab cheese from the grocery bag, I walk over to the refrigerator, open it up, neatly put cheese in the fridge, Close the fridge,
"actions": ['[WALK] <dining_room>', '[WALK] <food_food>', '[FIND] <food_food>', '[GRAB] <food_food>', '[WALK] <freezer>', '[OPEN] <freezer>', '[PUTIN] <food_food> <freezer>', '[CLOSE] <freezer>']
}
|
Scene_Objects:['water_glass', 'chair', 'cat', 'hanger', 'bathroom_counter', 'floor', 'bookshelf', 'keyboard', 'ceilinglamp', 'shower', 'foundation', 'couch', 'food_orange', 'powersocket', 'bench', 'clothes_gloves', 'toilet_paper', 'razor', 'food_donut', 'mouse', 'kitchen_counter', 'shoes', 'dishrack', 'dresser', 'pajamas', 'laundry_detergent', 'clothes_scarf', 'home_office', 'laser_pointer', 'creditcard']
Task:Drink
response:
| {
"thoughts": "Pour water into a glass and drink from the glass.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <water_glass>",
"[FIND] <water_glass>",
"[GRAB] <water_glass>",
"[WALK] <water>",
"[FIND] <water>",
"[GRAB] <water>",
"[POUR] <water> <water_glass>",
"[DRINK] <water_glass>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'tablelamp', 'chair', 'bathroom', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'ceilinglamp', 'deck_of_cards', 'balanceball', 'mousepad', 'dustpan', 'wall', 'nightstand', 'photoframe', 'wine', 'doorjamb', 'mouse', 'dresser', 'light', 'colander']
Task:Browse internet
response:
{
"thoughts": Go find a computer, turn it on, click google, and browse the internet.,
"actions": ['[WALK] <home_office>', '[WALK] <computer>', '[FIND] <computer>', '[SWITCHON] <computer>', '[TURNTO] <computer>', '[LOOKAT] <computer>']
}
|
Scene_Objects:['sheets', 'pot', 'chair', 'freezer', 'detergent', 'trashcan', 'dough', 'clothes_underwear', 'folder', 'stamp', 'bookshelf', 'clothes_jacket', 'table_cloth', 'mousepad', 'couch', 'food_jam', 'clothes_hat', 'pillow', 'nightstand', 'bench', 'oven', 'cutting_board', 'bed', 'food_food', 'check', 'doorjamb', 'console', 'cupboard', 'home_office', 'table']
Task:Pick up phone
response:
| {
"thoughts": "i go find the phone. i pick up the phone",
"actions": [
"[WALK] <home_office>",
"[WALK] <phone>",
"[FIND] <phone>",
"[GRAB] <phone>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['measuring_cup', 'chair', 'walllamp', 'trashcan', 'orchid', 'drawing', 'knifeblock', 'hanger', 'bathroom_counter', 'ceiling', 'bedroom', 'bookshelf', 'clothes_jacket', 'faucet', 'couch', 'clothes_pants', 'form', 'wall', 'cutting_board', 'photoframe', 'check', 'food_food', 'curtain', 'doorjamb', 'towel_rack', 'clothes_dress', 'door', 'sink', 'bathroom_cabinet', 'washing_machine']
Task:Wash clothes
response:
{
"thoughts": Walk to bathroom, open cabinet and collect clothes and put them into washing machine and wash clothes,
"actions": ['[WALK] <bathroom>', '[WALK] <bathroom_cabinet>', '[FIND] <bathroom_cabinet>', '[OPEN] <bathroom_cabinet>', '[FIND] <clothes_dress>', '[GRAB] <clothes_dress>', '[FIND] <washing_machine>', '[OPEN] <washing_machine>', '[PUTBACK] <clothes_dress> <washing_machine>', '[FIND] <clothes_pants>', '[GRAB] <clothes_pants>', '[PUTBACK] <clothes_pants> <washing_machine>', '[FIND] <clothes_jacket>', '[GRAB] <clothes_jacket>', '[PUTBACK] <clothes_jacket> <washing_machine>', '[CLOSE] <washing_machine>', '[SWITCHON] <washing_machine>']
}
|
Scene_Objects:['fork', 'toilet', 'freezer', 'ground_coffee', 'ceilingfan', 'milk', 'floor', 'ceiling', 'shampoo', 'purse', 'ceilinglamp', 'mat', 'faucet', 'oil', 'wall', 'pasta', 'tape', 'novel', 'food_food', 'food_donut', 'wooden_spoon', 'mouse', 'kitchen_counter', 'door', 'home_office', 'bathroom_cabinet', 'vase', 'soap', 'instrument_violin', 'creditcard']
Task:Put groceries in Fridge
response:
| {
"thoughts": "I walk in to the kitchen. I open the fridge. I place the groceries in the fridge. I close the fridge.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <freezer>",
"[FIND] <freezer>",
"[OPEN] <freezer>",
"[FIND] <food_food>",
"[GRAB] <food_food>",
"[PUTIN] <food_food> <freezer>",
"[CLOSE] <freezer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['water_glass', 'food_pizza', 'pot', 'microwave', 'bathroom', 'bills', 'knifeblock', 'cpuscreen', 'drawing', 'longboard', 'floor', 'phone', 'folder', 'bedroom', 'keyboard', 'mail', 'food_bacon', 'piano_bench', 'button', 'table_cloth', 'shower', 'couch', 'dining_room', 'wall', 'powersocket', 'cutting_board', 'water', 'tray', 'pencil', 'door']
Task:Drink
response:
{
"thoughts": Drinking a glass of water,
"actions": ['[WALK] <dining_room>', '[WALK] <water>', '[FIND] <water>', '[GRAB] <water>', '[FIND] <water_glass>', '[POUR] <water> <water_glass>', '[FIND] <water_glass>', '[GRAB] <water_glass>', '[DRINK] <water_glass>']
}
|
Scene_Objects:['filing_cabinet', 'food_ice_cream', 'crayon', 'bathroom', 'drawing', 'floor', 'phone', 'character', 'keyboard', 'remote_control', 'button', 'ceilinglamp', 'shower', 'window', 'foundation', 'mousepad', 'couch', 'pillow', 'wall', 'oven', 'check', 'food_food', 'television', 'dresser', 'sink', 'table', 'dishwasher', 'clothes_skirt', 'tooth_paste', 'dirt']
Task:Change TV channel
response:
| {
"thoughts": "go to living room. grab remote control. sit on couch. point remote at tv. push buttons.",
"actions": [
"[WALK] <home_office>",
"[WALK] <remote_control>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[FIND] <couch>",
"[SIT] <couch>",
"[TURNTO] <remote_control>",
"[POINTAT] <remote_control>",
"[FIND] <button>",
"[PUSH] <button>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['water_glass', 'microwave', 'needle', 'kitchen_cabinet', 'desk', 'alcohol', 'mat', 'ceilinglamp', 'balanceball', 'window', 'couch', 'clothes_pants', 'dining_room', 'nightstand', 'wall', 'wall_clock', 'oven', 'toilet_paper', 'mouthwash', 'water', 'ice', 'tray', 'rag', 'doorjamb', 'towel_rack', 'coffee_pot', 'pencil', 'door', 'dishwasher', 'vase']
Task:Drink
response:
{
"thoughts": Get a glass of water to drink,
"actions": ['[WALK] <dining_room>', '[WALK] <kitchen_cabinet>', '[FIND] <kitchen_cabinet>', '[FIND] <kitchen_cabinet>', '[FIND] <water>', '[GRAB] <water>', '[FIND] <water_glass>', '[POUR] <water> <water_glass>', '[GRAB] <water_glass>', '[DRINK] <water_glass>']
}
|
Scene_Objects:['tablelamp', 'scissors', 'chair', 'freezer', 'dog', 'trashcan', 'ceilingfan', 'drawing', 'bills', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'bedroom', 'keyboard', 'bookshelf', 'computer', 'ceilinglamp', 'balanceball', 'faucet', 'mousepad', 'wall', 'nightstand', 'television', 'doorjamb', 'cards', 'mouse', 'light', 'table']
Task:Browse internet
response:
| {
"thoughts": "I go to my desk. I turn on the computer. I open a web page and look at it.",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>",
"[TURNTO] <computer>",
"[LOOKAT] <computer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'tablelamp', 'chair', 'walllamp', 'drawing', 'orchid', 'cpuscreen', 'wallshelf', 'hanger', 'floor', 'phone', 'ceiling', 'desk', 'keyboard', 'computer', 'mat', 'ceilinglamp', 'tvstand', 'mousepad', 'couch', 'pillow', 'wall', 'nightstand', 'oven', 'powersocket', 'television', 'tray', 'doorjamb', 'towel_rack', 'mouse', 'dresser', 'door', 'light']
Task:Surf web for money legitimate making opportunities
response:
{
"thoughts": Sit down in front of computer. Turn computer on. Go online. Surf the web.,
"actions": ['[WALK] <home_office>', '[WALK] <desk>', '[FIND] <chair>', '[SIT] <chair>', '[FIND] <computer>', '[SWITCHON] <computer>', '[TURNTO] <computer>', '[LOOKAT] <computer>', '[FIND] <keyboard>', '[TYPE] <keyboard>']
}
|
Scene_Objects:['tablelamp', 'food_ice_cream', 'measuring_cup', 'freezer', 'ceilingfan', 'cpuscreen', 'longboard', 'ceiling', 'character', 'bedroom', 'toaster', 'stereo', 'balanceball', 'oil', 'couch', 'food_orange', 'pillow', 'food_kiwi', 'nightstand', 'photoframe', 'coffe_maker', 'rag', 'doorjamb', 'towel_rack', 'shoes', 'controller', 'napkin', 'bathroom_cabinet', 'laser_pointer', 'dirt']
Task:Pick up phone
response:
| {
"thoughts": "Go to the living room. Find the telephone. Lift up the receiver.",
"actions": [
"[WALK] <home_office>",
"[WALK] <phone>",
"[FIND] <phone>",
"[GRAB] <phone>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'tablelamp', 'toilet', 'chair', 'walllamp', 'detergent', 'trashcan', 'food_carrot', 'knifeblock', 'drawing', 'cpuscreen', 'knife', 'bathroom_counter', 'desk', 'keyboard', 'bookshelf', 'table_cloth', 'toaster', 'ceilinglamp', 'faucet', 'mousepad', 'pillow', 'dining_room', 'oven', 'photoframe', 'food_food', 'kitchen_counter', 'dresser', 'sink', 'home_office']
Task:Turn off light
response:
{
"thoughts": Walk to the lamp. Find the switch. Click the switch off.,
"actions": ['[WALK] <bedroom>', '[WALK] <floor_lamp>', '[SWITCHOFF] <floor_lamp>']
}
|
Scene_Objects:['mop', 'bathroom', 'ceilingfan', 'drawing', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'remote_control', 'computer', 'clothes_jacket', 'toaster', 'ceilinglamp', 'tvstand', 'balanceball', 'window', 'food_noodles', 'couch', 'clothes_pants', 'pillow', 'wall', 'oven', 'check', 'television', 'tray', 'curtain', 'sink', 'table', 'light', 'teeth']
Task:Watch TV
response:
| {
"thoughts": "You walk to the living room and sit on the sofa. You grab the remote control, look at it and point at the television, you watch television.",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[SIT] <couch>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[TURNTO] <remote_control>",
"[LOOKAT] <remote_control>",
"[FIND] <television>",
"[TURNTO] <television>",
"[POINTAT] <television>",
"[WATCH] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['dry_pasta', 'pot', 'food_salt', 'jelly', 'walllamp', 'detergent', 'knife', 'drawing', 'hanger', 'floor', 'dvd_player', 'character', 'desk', 'bedroom', 'plate', 'dish_soap', 'oil', 'wall', 'powersocket', 'bed', 'light_bulb', 'coffe_maker', 'toilet_paper', 'rag', 'doorjamb', 'kitchen_counter', 'cupboard', 'sink', 'spoon', 'dishwasher']
Task:Wash dishes with dishwasher
response:
{
"thoughts": Put dishes in the dishwasher and start the dishwasher.,
"actions": ['[RUN] <dining_room>', '[WALK] <bowl>', '[FIND] <bowl>', '[GRAB] <bowl>', '[WALK] <dishwasher>', '[OPEN] <dishwasher>', '[PUTBACK] <bowl> <dishwasher>', '[FIND] <plate>', '[GRAB] <plate>', '[WALK] <dishwasher>', '[PUTBACK] <plate> <dishwasher>', '[FIND] <dish_soap>', '[GRAB] <dish_soap>', '[WALK] <dishwasher>', '[PUTBACK] <dish_soap> <dishwasher>', '[CLOSE] <dishwasher>', '[SWITCHON] <dishwasher>']
}
|
Scene_Objects:['pot', 'chair', 'freezer', 'walllamp', 'trashcan', 'knifeblock', 'needle', 'cpuscreen', 'floor', 'character', 'keyboard', 'table_cloth', 'plate', 'faucet', 'window', 'oil', 'mousepad', 'pillow', 'wall', 'oven', 'food_cheese', 'bed', 'tray', 'kettle', 'kitchen_counter', 'dresser', 'cupboard', 'sink', 'home_office', 'light']
Task:Wash dishes by hand
response:
| {
"thoughts": "Walking towards sink and look at the leftover plates. Scrape dishes to remove leftover food - use a rubber spatula or paper towel. Fill sink or dishpan with clean, hot water. Then wash and rinse to removing residues. Finally leave it to dry.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <kitchen_counter>",
"[FIND] <plate>",
"[TURNTO] <plate>",
"[LOOKAT] <plate>",
"[GRAB] <plate>",
"[SCRUB] <plate>",
"[RINSE] <plate>",
"[WASH] <plate>",
"[PUTBACK] <plate> <kitchen_counter>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['fork', 'microwave', 'ceilingfan', 'drawing', 'hanger', 'bathroom_counter', 'floor', 'ceiling', 'food_cake', 'thread', 'faucet', 'towel', 'dining_room', 'wall', 'nightstand', 'bench', 'photoframe', 'water', 'rag', 'lighter', 'cards', 'console', 'mouse', 'conditioner', 'sink', 'bathroom_cabinet', 'table', 'soap', 'laser_pointer', 'hands_both']
Task:Wash hands
response:
{
"thoughts": Go into the bathroom and go over to the sink. Grab soap and turn on the water. Wet your hands and rub them together so that the soap lathers. Rub the soap all over your hands thoroughly. Rinse off the soap with water and turn off the sink. Dry your hands with a towel.,
"actions": ['[WALK] <bathroom>', '[WALK] <sink>', '[FIND] <soap>', '[GRAB] <soap>', '[FIND] <water>', '[GRAB] <water>', '[FIND] <hands_both>', '[POUR] <water> <hands_both>', '[FIND] <hands_both>', '[SCRUB] <hands_both>', '[PUTOBJBACK] <soap>', '[RINSE] <hands_both>', '[FIND] <towel>', '[GRAB] <towel>', '[WIPE] <hands_both>']
}
|
Scene_Objects:['tablelamp', 'chair', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'purse', 'computer', 'ceilinglamp', 'mat', 'balanceball', 'electric_shaver', 'window', 'mousepad', 'dustpan', 'pillow', 'wall', 'nightstand', 'photoframe', 'food_food', 'doorjamb', 'mouse', 'door', 'light']
Task:Browse internet
response:
| {
"thoughts": "turn on the computer and browse the internet",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sponge', 'water_glass', 'pot', 'glue', 'walllamp', 'trashcan', 'knifeblock', 'ceiling', 'clothes_jacket', 'stovefan', 'toaster', 'mat', 'faucet', 'microphone', 'wall', 'cup', 'bed', 'food_food', 'check', 'water', 'food_donut', 'curtain', 'kitchen_counter', 'chessboard', 'cupboard', 'laundry_detergent', 'sink', 'spoon', 'bathroom_cabinet', 'dirt']
Task:Drink
response:
{
"thoughts": I walk to the kitchen. open the cupboard. Find the water glass, grab it. Fill the glass with water and drink water.,
"actions": ['[WALK] <dining_room>', '[WALK] <cupboard>', '[FIND] <cupboard>', '[OPEN] <cupboard>', '[FIND] <water_glass>', '[GRAB] <water_glass>', '[WALK] <water>', '[FIND] <water>', '[GRAB] <water>', '[POUR] <water> <water_glass>', '[DRINK] <water_glass>']
}
|
Scene_Objects:['filing_cabinet', 'tablelamp', 'tea', 'toilet', 'chair', 'detergent', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'mail', 'ceilinglamp', 'balanceball', 'mousepad', 'wall', 'nightstand', 'novel', 'food_onion', 'doorjamb', 'mouse', 'napkin', 'light', 'clothes_skirt']
Task:Browse internet
response:
| {
"thoughts": "I go to the living room, where my computer is. I locate it, switch it on and type desired URL adress in the web browser.",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>",
"[FIND] <keyboard>",
"[TYPE] <keyboard>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['chair', 'walllamp', 'drawing', 'knife', 'wallshelf', 'orchid', 'hanger', 'bathroom_counter', 'floor', 'ceiling', 'desk', 'remote_control', 'computer', 'mat', 'tvstand', 'ceilinglamp', 'oil', 'closetdrawer', 'couch', 'pillow', 'wall', 'box', 'photoframe', 'television', 'coffee_filter', 'after_shave', 'doorjamb', 'diary', 'kitchen_counter', 'dresser']
Task:Change TV channel
response:
{
"thoughts": Walk to livingroom, grab remote and change the channel on television.,
"actions": ['[WALK] <home_office>', '[WALK] <remote_control>', '[FIND] <remote_control>', '[GRAB] <remote_control>', '[FIND] <television>', '[SWITCHON] <television>', '[PUTOBJBACK] <remote_control>']
}
|
Scene_Objects:['tablelamp', 'food_ice_cream', 'chair', 'nail_polish', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'character', 'desk', 'keyboard', 'bookshelf', 'computer', 'button', 'ceilinglamp', 'balanceball', 'shower', 'mousepad', 'pillow', 'wall', 'nightstand', 'basket_for_clothes', 'toothbrush_holder', 'doorjamb', 'mouse', 'light', 'vase']
Task:Browse internet
response:
| {
"thoughts": "Walk to the Home Office. Find the computer. Turn on the computer.",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'tablelamp', 'microwave', 'toothbrush', 'drawing', 'orchid', 'wallshelf', 'kitchen_cabinet', 'floor', 'ceiling', 'desk', 'bedroom', 'stovefan', 'toaster', 'ceilinglamp', 'faucet', 'shower', 'mousepad', 'pillow', 'wall', 'pasta', 'oven', 'coffe_maker', 'water', 'tray', 'towel_rack', 'kitchen_counter', 'headset', 'sauce_pan', 'cupboard']
Task:Cook some food
response:
{
"thoughts": Cook some pasta on the stove,
"actions": ['[WALK] <dining_room>', '[WALK] <kitchen_cabinet>', '[FIND] <kitchen_cabinet>', '[OPEN] <kitchen_cabinet>', '[FIND] <pasta>', '[GRAB] <pasta>', '[FIND] <water>', '[GRAB] <water>', '[FIND] <sauce_pan>', '[POUR] <water> <sauce_pan>', '[FIND] <oven>', '[SWITCHON] <oven>', '[PUTBACK] <pasta> <sauce_pan>', '[FIND] <sauce_pan>', '[GRAB] <sauce_pan>', '[PUTBACK] <sauce_pan> <oven>']
}
|
Scene_Objects:['tablelamp', 'juice', 'toilet', 'chair', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'hanger', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'toaster', 'ceilinglamp', 'stovefan', 'balanceball', 'mat', 'mousepad', 'oil', 'wall', 'nightstand', 'photoframe', 'doorjamb', 'mouse', 'light', 'dishwasher']
Task:Work
response:
| {
"thoughts": "Go to office, sit at desk, turn on computer, enter password, open application and begin work",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>",
"[FIND] <keyboard>",
"[TYPE] <keyboard>",
"[FIND] <mouse>",
"[PUSH] <mouse>",
"[TURNTO] <computer>",
"[LOOKAT] <computer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'crayon', 'toilet', 'chair', 'bathroom', 'orchid', 'floor', 'ceiling', 'character', 'bedroom', 'desk', 'computer', 'tvstand', 'faucet', 'stereo', 'food_orange', 'photoframe', 'novel', 'television', 'curtain', 'doorjamb', 'diary', 'towel_rack', 'mouse', 'shoes', 'dresser', 'cupboard', 'sink', 'bathtub', 'home_office']
Task:Get dressed
response:
{
"thoughts": I get dressed in my bedroom in the mornings.,
"actions": ['[WALK] <bedroom>', '[WALK] <shoes>', '[FIND] <shoes>', '[GRAB] <shoes>', '[PUTON] <shoes>']
}
|
Scene_Objects:['sponge', 'sheets', 'filing_cabinet', 'food_bread', 'toilet', 'ground_coffee', 'drawing', 'cpuscreen', 'bathroom_counter', 'floor', 'ceiling', 'desk', 'keyboard', 'alcohol', 'table_cloth', 'tvstand', 'balanceball', 'shower', 'plate', 'iron', 'mousepad', 'wall', 'oven', 'photoframe', 'light_bulb', 'coffe_maker', 'centerpiece', 'mouse', 'dresser', 'maindoor']
Task:Go to toilet
response:
| {
"thoughts": "I walk to the entrance of the bathroom. I turn to the door. I open the door to the bathroom. I walk inside of the bathroom. I find the toilet inside of the bathroom and walk over to it. I turn and look at the toilet.",
"actions": [
"[WALK] <bathroom>",
"[WALK] <toilet>",
"[FIND] <toilet>",
"[TURNTO] <toilet>",
"[LOOKAT] <toilet>",
"[WALK] <toilet>",
"[TURNTO] <toilet>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['microwave', 'toilet', 'chair', 'freezer', 'detergent', 'trashcan', 'drawing', 'wallshelf', 'clothes_underwear', 'floor', 'ceiling', 'phone', 'computer', 'toaster', 'mat', 'tvstand', 'shower', 'closetdrawer', 'wall', 'oven', 'cup', 'novel', 'water', 'drinking_glass', 'doorjamb', 'mouse', 'kitchen_counter', 'comb', 'cupboard', 'door']
Task:Drink
response:
{
"thoughts": go to the kitchen,
"actions": ['[WALK] <dining_room>', '[WALK] <freezer>', '[FIND] <freezer>', '[OPEN] <freezer>', '[FIND] <drinking_glass>', '[GRAB] <drinking_glass>', '[FIND] <water>', '[DRINK] <drinking_glass>']
}
|
Scene_Objects:['pot', 'freezer', 'ceilingfan', 'dough', 'milk', 'knifeblock', 'bills', 'floor', 'ceiling', 'bedroom', 'keyboard', 'bookshelf', 'computer', 'mop_bucket', 'ceilinglamp', 'microphone', 'wall', 'bench', 'powersocket', 'clothes_gloves', 'light_bulb', 'food_food', 'coffe_maker', 'tray', 'curtain', 'shaving_cream', 'controller', 'door', 'maindoor', 'dirt']
Task:Put groceries in Fridge
response:
| {
"thoughts": "Go to kitchen and put groceries in fridge",
"actions": [
"[WALK] <dining_room>",
"[WALK] <food_food>",
"[FIND] <food_food>",
"[GRAB] <food_food>",
"[WALK] <freezer>",
"[OPEN] <freezer>",
"[PUTIN] <food_food> <freezer>",
"[CLOSE] <freezer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['spectacles', 'sheets', 'filing_cabinet', 'food_salt', 'toilet', 'band-aids', 'walllamp', 'wallshelf', 'cpuscreen', 'hanger', 'floor', 'phone', 'computer', 'mop_bucket', 'stovefan', 'tvstand', 'deck_of_cards', 'closetdrawer', 'newspaper', 'coffe_maker', 'doorjamb', 'dresser', 'cupboard', 'door', 'food_steak', 'bag', 'sink', 'table', 'dishwasher', 'bathroom_cabinet']
Task:Put umbrella away
response:
{
"thoughts": Place the umbrella in the umbrella bucket point first.,
"actions": ['[WALK] <home_office>', '[WALK] <mop_bucket>']
}
|
Scene_Objects:['filing_cabinet', 'juice', 'freezer', 'blow_dryer', 'trashcan', 'drawing', 'cpuscreen', 'longboard', 'floor', 'bookmark', 'ceilinglamp', 'mat', 'window', 'couch', 'microphone', 'clothes_hat', 'pillow', 'wall', 'bench', 'cup', 'keys', 'television', 'towel_rack', 'cd', 'cupboard', 'conditioner', 'door', 'home_office', 'table', 'vase']
Task:Relax on sofa
response:
| {
"thoughts": "Go to the living room and relax on the sofa with a book",
"actions": [
"[RUN] <home_office>",
"[WALK] <novel>",
"[FIND] <novel>",
"[GRAB] <novel>",
"[WALK] <couch>",
"[SIT] <couch>",
"[TURNTO] <novel>",
"[LOOKAT] <novel>",
"[FIND] <bookmark>",
"[GRAB] <bookmark>",
"[DROP] <bookmark>",
"[TURNTO] <novel>",
"[LOOKAT] <novel>",
"[READ] <novel>",
"[READ] <novel>",
"[READ] <novel>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['microwave', 'toilet', 'walllamp', 'bathroom', 'orchid', 'wallshelf', 'bathroom_counter', 'floor', 'phone', 'ceiling', 'desk', 'bookshelf', 'keyboard', 'computer', 'stovefan', 'tvstand', 'faucet', 'plate', 'mousepad', 'wall', 'powersocket', 'bench', 'oven', 'cutting_board', 'tape', 'cup', 'doorjamb', 'sink', 'light', 'table']
Task:Pick up phone
response:
{
"thoughts": walk over and pick up the phone,
"actions": ['[WALK] <phone>', '[GRAB] <phone>']
}
|
Scene_Objects:['food_pizza', 'tablelamp', 'toilet', 'chair', 'walllamp', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'ceilinglamp', 'balanceball', 'faucet', 'mousepad', 'wall', 'nightstand', 'cup', 'coffe_maker', 'doorjamb', 'mouse', 'chessboard', 'controller', 'bathroom_cabinet', 'light']
Task:Write an email
response:
| {
"thoughts": "enter office, find computer, turn it on,",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[FIND] <computer>",
"[SWITCHON] <computer>",
"[FIND] <keyboard>",
"[TOUCH] <keyboard>",
"[TYPE] <keyboard>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sponge', 'pot', 'filing_cabinet', 'sheets', 'detergent', 'milk', 'hanger', 'phone', 'desk', 'clothes_jacket', 'plate', 'dish_soap', 'mousepad', 'couch', 'candle', 'pillow', 'bowl', 'basket_for_clothes', 'standingmirror', 'keys', 'food_food', 'check', 'razor', 'coffee_filter', 'mouse', 'kitchen_counter', 'controller', 'door', 'bathtub', 'dishwasher']
Task:Wash dishes with dishwasher
response:
{
"thoughts": Put dishes in the dishwasher and start the dishwasher.,
"actions": ['[RUN] <dining_room>', '[WALK] <bowl>', '[FIND] <bowl>', '[GRAB] <bowl>', '[WALK] <dishwasher>', '[OPEN] <dishwasher>', '[PUTBACK] <bowl> <dishwasher>', '[FIND] <plate>', '[GRAB] <plate>', '[WALK] <dishwasher>', '[PUTBACK] <plate> <dishwasher>', '[FIND] <dish_soap>', '[GRAB] <dish_soap>', '[WALK] <dishwasher>', '[PUTBACK] <dish_soap> <dishwasher>', '[CLOSE] <dishwasher>', '[SWITCHON] <dishwasher>']
}
|
Scene_Objects:['filing_cabinet', 'walllamp', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'desk', 'bookshelf', 'keyboard', 'computer', 'button', 'mat', 'balanceball', 'window', 'candle', 'wall', 'powersocket', 'photoframe', 'novel', 'coffee_filter', 'ice', 'doorjamb', 'kettle', 'console', 'door', 'bathroom_cabinet', 'light', 'home_office', 'table', 'soap']
Task:Turn on light
response:
| {
"thoughts": "I walk to wall. I put my hand on the switch. I push the button. I turn on the lights.",
"actions": [
"[WALK] <light>",
"[FIND] <light>",
"[SWITCHON] <light>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['walllamp', 'drawing', 'orchid', 'wallshelf', 'cpuscreen', 'floor', 'ceiling', 'bookshelf', 'remote_control', 'computer', 'mat', 'tvstand', 'ceilinglamp', 'shower', 'window', 'couch', 'pillow', 'wall', 'powersocket', 'bench', 'photoframe', 'check', 'television', 'curtain', 'doorjamb', 'laundry_detergent', 'door', 'table', 'light', 'soap']
Task:Watch TV
response:
{
"thoughts": turn tv on, watch TV,
"actions": ['[WALK] <home_office>', '[WALK] <couch>', '[FIND] <couch>', '[SIT] <couch>', '[FIND] <remote_control>', '[TURNTO] <remote_control>', '[FIND] <television>', '[TURNTO] <television>', '[WATCH] <television>']
}
|
Scene_Objects:['food_bread', 'scissors', 'toilet', 'freezer', 'ground_coffee', 'ceilingfan', 'knifeblock', 'orchid', 'hanger', 'floor', 'bedroom', 'tea_bag', 'envelope', 'clothes_jacket', 'toaster', 'balanceball', 'faucet', 'iron', 'bed', 'coffe_maker', 'wooden_spoon', 'cards', 'headset', 'dresser', 'napkin', 'spoon', 'vase', 'instrument_violin', 'laser_pointer', 'creditcard']
Task:Receive credit card
response:
| {
"thoughts": "hear the dooble, go to door, take de envelope, open it with a scissor, take the creditcard",
"actions": [
"[WALK] <home_office>",
"[WALK] <envelope>",
"[FIND] <envelope>",
"[GRAB] <envelope>",
"[WALK] <home_office>",
"[WALK] <scissors>",
"[FIND] <scissors>",
"[OPEN] <envelope>",
"[FIND] <creditcard>",
"[GRAB] <creditcard>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'tablelamp', 'beer', 'detergent', 'food_butter', 'orchid', 'hanger', 'floor', 'ceiling', 'computer', 'toaster', 'mat', 'closetdrawer', 'broom', 'wall', 'nightstand', 'toy', 'newspaper', 'toilet_paper', 'television', 'wine', 'doorjamb', 'kitchen_counter', 'shoes', 'chessboard', 'dresser', 'door', 'dishwasher', 'instrument_violin', 'laser_pointer']
Task:Put away shoes
response:
{
"thoughts": Pick up shoes from entry hall. Open entry way closet. Put shoes in closet. Close closet door.,
"actions": ['[WALK] <home_office>', '[WALK] <shoes>', '[FIND] <shoes>', '[GRAB] <shoes>', '[FIND] <dresser>', '[OPEN] <dresser>', '[PUTIN] <shoes> <dresser>', '[CLOSE] <dresser>']
}
|
Scene_Objects:['tea', 'jelly', 'dog', 'walllamp', 'bathroom_counter', 'thread', 'table_cloth', 'tvstand', 'balanceball', 'shower', 'window', 'oil', 'microphone', 'clothes_hat', 'nightstand', 'photoframe', 'bed', 'check', 'coffe_maker', 'toothbrush_holder', 'kitchen_counter', 'dresser', 'cupboard', 'controller', 'laundry_detergent', 'maindoor', 'sink', 'bathroom_cabinet', 'vase', 'soap']
Task:Change TV channel
response:
| {
"thoughts": "take remote control, push \"change channel\" button",
"actions": [
"[WALK] <home_office>",
"[WALK] <remote_control>",
"[FIND] <remote_control>",
"[PUSH] <remote_control>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['crayon', 'needle', 'clothes_underwear', 'phone', 'desk', 'stamp', 'clothes_jacket', 'stovefan', 'mat', 'tvstand', 'window', 'towel', 'pillow', 'cutting_board', 'basket_for_clothes', 'cup', 'bed', 'photoframe', 'newspaper', 'food_food', 'mouthwash', 'television', 'wine', 'kettle', 'dresser', 'bathtub', 'bathroom_cabinet', 'table', 'soap', 'washing_machine']
Task:Wash clothes
response:
{
"thoughts": Walk to the kitchen and find the basket of clothes. Put the soap and clothes into the washing machine. Turn on the washing machine.,
"actions": ['[WALK] <dining_room>', '[WALK] <basket_for_clothes>', '[FIND] <basket_for_clothes>', '[FIND] <washing_machine>', '[TURNTO] <washing_machine>', '[FIND] <soap>', '[GRAB] <soap>', '[OPEN] <washing_machine>', '[PUTBACK] <soap> <washing_machine>', '[FIND] <clothes_jacket>', '[GRAB] <clothes_jacket>', '[PUTBACK] <clothes_jacket> <washing_machine>', '[CLOSE] <washing_machine>', '[SWITCHON] <washing_machine>']
}
|
Scene_Objects:['dry_pasta', 'pot', 'food_pizza', 'food_ice_cream', 'freezer', 'walllamp', 'knife', 'bathroom_counter', 'floor', 'stamp', 'tea_bag', 'keyboard', 'food_cake', 'bookmark', 'toaster', 'plate', 'stereo', 'dish_soap', 'food_noodles', 'wall', 'oven', 'bench', 'coffe_maker', 'food_food', 'tray', 'food_onion', 'dishrack', 'conditioner', 'bathroom_cabinet', 'dishwasher']
Task:Wash dishes with dishwasher
response:
| {
"thoughts": "Walk to kitchen. Find dishwasher. Open dishwasher. Find dishrack. Find plate 1. Grab plate 1. Put plate 1 in dishrack. Repeat for dish 2. Find dish soap. Grab dish soap. Put dish soap to dishwasher. Close dishwasher. Switch on dishwasher.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <dishwasher>",
"[FIND] <dishwasher>",
"[OPEN] <dishwasher>",
"[FIND] <dishrack>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[PUTBACK] <plate> <dishrack>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[PUTBACK] <plate> <dishrack>",
"[FIND] <dish_soap>",
"[GRAB] <dish_soap>",
"[PUTBACK] <dish_soap> <dishwasher>",
"[CLOSE] <dishwasher>",
"[SWITCHON] <dishwasher>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sheets', 'toilet', 'freezer', 'cat', 'orchid', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'bookshelf', 'computer', 'stovefan', 'ceilinglamp', 'tvstand', 'balanceball', 'oil', 'couch', 'clothes_hat', 'wall', 'basket_for_clothes', 'food_food', 'coffe_maker', 'curtain', 'doorjamb', 'towel_rack', 'comb', 'dresser', 'maindoor', 'soap']
Task:Put groceries in Fridge
response:
{
"thoughts": I put my groceries into the fridge.,
"actions": ['[WALK] <food_food>', '[GRAB] <food_food>', '[FIND] <freezer>', '[OPEN] <freezer>', '[PUTIN] <food_food> <freezer>', '[CLOSE] <freezer>']
}
|
Scene_Objects:['filing_cabinet', 'toilet', 'ceilingfan', 'food_egg', 'hanger', 'ceiling', 'dvd_player', 'character', 'thread', 'clothes_jacket', 'toaster', 'deck_of_cards', 'foundation', 'mousepad', 'oil', 'pillow', 'oven', 'toy', 'food_food', 'check', 'lighter', 'centerpiece', 'dishrack', 'coffee', 'hair', 'laundry_detergent', 'maindoor', 'sink', 'door', 'teeth']
Task:Turn on light
response:
| {
"thoughts": "I walk into my bedroom, go to the lamp and use the switch to turn it on.",
"actions": [
"[WALK] <bedroom>",
"[WALK] <floor_lamp>",
"[FIND] <floor_lamp>",
"[TURNTO] <floor_lamp>",
"[LOOKAT] <floor_lamp>",
"[WALK] <floor_lamp>",
"[SWITCHON] <floor_lamp>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'chair', 'walllamp', 'knife', 'desk', 'bookshelf', 'button', 'clothes_jacket', 'stovefan', 'deck_of_cards', 'mousepad', 'closetdrawer', 'oil', 'towel', 'pillow', 'broom', 'basket_for_clothes', 'photoframe', 'novel', 'coffe_maker', 'television', 'wine', 'curtain', 'after_shave', 'cards', 'home_office', 'light', 'clothes_skirt', 'soap', 'washing_machine']
Task:Wash clothes
response:
{
"thoughts": walk to the bathroom, grab the basket for clothes, find soap, put in washing machine, turn on,
"actions": ['[WALK] <bathroom>', '[WALK] <basket_for_clothes>', '[FIND] <basket_for_clothes>', '[GRAB] <basket_for_clothes>', '[WALK] <washing_machine>', '[OPEN] <washing_machine>', '[PUTBACK] <basket_for_clothes> <washing_machine>', '[FIND] <soap>', '[GRAB] <soap>', '[WALK] <washing_machine>', '[PUTBACK] <soap> <washing_machine>', '[CLOSE] <washing_machine>', '[SWITCHON] <washing_machine>']
}
|
Scene_Objects:['sheets', 'microwave', 'chair', 'freezer', 'food_butter', 'drawing', 'food_egg', 'floor', 'ceiling', 'character', 'stovefan', 'ceilinglamp', 'faucet', 'shower', 'couch', 'wall', 'nightstand', 'bench', 'oven', 'cup', 'novel', 'ice', 'curtain', 'doorjamb', 'kettle', 'pencil', 'headset', 'sink', 'bathroom_cabinet', 'light']
Task:Drink
response:
| {
"thoughts": "Walk into kitchen, look for cup, walk to cup, look for refrigerator, walk to refrigerator, open door, grab water, pour water into cup",
"actions": [
"[WALK] <dining_room>",
"[WALK] <cup>",
"[FIND] <cup>",
"[TURNTO] <cup>",
"[LOOKAT] <cup>",
"[WALK] <cup>",
"[FIND] <freezer>",
"[TURNTO] <freezer>",
"[LOOKAT] <freezer>",
"[WALK] <freezer>",
"[GRAB] <cup>",
"[WALK] <water>",
"[FIND] <water>",
"[GRAB] <water>",
"[POUR] <water> <cup>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['toilet', 'bathroom', 'trashcan', 'cpuscreen', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'desk', 'bookshelf', 'alcohol', 'mat', 'electric_shaver', 'oil', 'wall', 'powersocket', 'box', 'food_cheese', 'photoframe', 'bed', 'food_food', 'coffe_maker', 'doorjamb', 'lighter', 'dishrack', 'door', 'bathroom_cabinet', 'light', 'dishwasher', 'coffee']
Task:Turn on light
response:
{
"thoughts": push any ligth button to turn it on, for example in the living room,
"actions": ['[WALK] <home_office>', '[WALK] <light>', '[FIND] <light>', '[SWITCHON] <light>']
}
|
Scene_Objects:['glue', 'pot', 'chair', 'band-aids', 'detergent', 'drawing', 'needle', 'hanger', 'floor', 'ceiling', 'phone', 'desk', 'computer', 'stovefan', 'ceilinglamp', 'tvstand', 'towel', 'dining_room', 'wall', 'basket_for_clothes', 'novel', 'after_shave', 'doorjamb', 'centerpiece', 'headset', 'dresser', 'cupboard', 'door', 'light', 'scrabble']
Task:Turn on light
response:
| {
"thoughts": "Switch on light in home office",
"actions": [
"[WALK] <home_office>",
"[WALK] <light>",
"[FIND] <light>",
"[PLUGIN] <light>",
"[SWITCHON] <light>",
"[WALK] <home_office>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_salt', 'toilet', 'walllamp', 'food_carrot', 'knife', 'wallshelf', 'thread', 'envelope', 'computer', 'stovefan', 'ceilinglamp', 'shower', 'couch', 'pillow', 'broom', 'wall', 'bench', 'powersocket', 'tape', 'shelf', 'basket_for_clothes', 'photoframe', 'bed', 'oven_mitts', 'towel_rack', 'shaving_cream', 'kitchen_counter', 'shoes', 'table', 'shoe_rack']
Task:Listen to music
response:
{
"thoughts": I walk to the living room. I find a radio there and switch it on to listen to music.,
"actions": ['[WALK] <home_office>', '[WALK] <stereo>', '[FIND] <stereo>', '[SWITCHON] <stereo>']
}
|
Scene_Objects:['scissors', 'microwave', 'toilet', 'freezer', 'ceilingfan', 'drawing', 'knifeblock', 'food_egg', 'cpuscreen', 'clothes_underwear', 'tea_bag', 'keyboard', 'faucet', 'clothes_socks', 'food_noodles', 'dining_room', 'bench', 'cup', 'bed', 'check', 'toilet_paper', 'television', 'after_shave', 'console', 'comb', 'hairbrush', 'cupboard', 'bathroom_cabinet', 'board_game', 'teeth']
Task:Pet cat
response:
| {
"thoughts": "I walk to my cat, and I pet him",
"actions": [
"[WALK] <cat>",
"[TOUCH] <cat>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'chair', 'walllamp', 'drawing', 'orchid', 'cpuscreen', 'wallshelf', 'hanger', 'clothes_underwear', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'computer', 'button', 'mat', 'ceilinglamp', 'tvstand', 'mousepad', 'couch', 'pillow', 'wall', 'nightstand', 'oven', 'television', 'doorjamb', 'mouse', 'dresser', 'door']
Task:Work
response:
{
"thoughts": Go to the computer, sit in the chair, pull up computer screen get the mouse and keyboard load work software,
"actions": ['[WALK] <home_office>', '[WALK] <chair>', '[FIND] <chair>', '[SIT] <chair>', '[FIND] <computer>', '[SWITCHON] <computer>', '[TURNTO] <computer>', '[LOOKAT] <computer>', '[FIND] <mouse>', '[GRAB] <mouse>', '[FIND] <mousepad>', '[PUTBACK] <mouse> <mousepad>', '[TURNTO] <computer>', '[POINTAT] <computer>', '[PULL] <mouse>', '[FIND] <button>', '[PUSH] <button>', '[GRAB] <mouse>', '[PUTOBJBACK] <mouse>', '[FIND] <keyboard>', '[TYPE] <keyboard>']
}
|
Scene_Objects:['pot', 'food_bread', 'ceilingfan', 'drawing', 'knife', 'longboard', 'floor', 'ceiling', 'keyboard', 'remote_control', 'ceilinglamp', 'tvstand', 'balanceball', 'laptop', 'window', 'foundation', 'clothes_socks', 'couch', 'dining_room', 'broom', 'wall', 'cup', 'coffee_table', 'television', 'toothbrush_holder', 'curtain', 'lighter', 'mouse', 'maindoor', 'table']
Task:Watch TV
response:
| {
"thoughts": "Sit down, turn on the TV with the remote control that is on the coffee table.",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[SIT] <couch>",
"[FIND] <coffee_table>",
"[TURNTO] <coffee_table>",
"[LOOKAT] <coffee_table>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[FIND] <television>",
"[TURNTO] <television>",
"[LOOKAT] <television>",
"[TURNTO] <television>",
"[POINTAT] <television>",
"[SWITCHON] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['toilet', 'ceilingfan', 'drawing', 'knife', 'cpuscreen', 'floor', 'ceiling', 'desk', 'remote_control', 'button', 'toaster', 'ceilinglamp', 'tvstand', 'balanceball', 'window', 'couch', 'food_orange', 'dining_room', 'wall', 'tape', 'photoframe', 'television', 'tray', 'curtain', 'kitchen_counter', 'vacuum_cleaner', 'sink', 'instrument_guitar', 'table', 'bathroom_cabinet']
Task:Watch TV
response:
{
"thoughts": Walk to living room. Find remote control. Find television. Point remote at television. Press "on" button. Watch.,
"actions": ['[WALK] <home_office>', '[WALK] <television>', '[FIND] <television>', '[FIND] <remote_control>', '[TURNTO] <television>', '[POINTAT] <television>', '[FIND] <button>', '[PUSH] <button>', '[TURNTO] <television>', '[WATCH] <television>']
}
|
Scene_Objects:['crayon', 'chair', 'ground_coffee', 'trashcan', 'dough', 'toothbrush', 'knifeblock', 'hanger', 'floor', 'ceiling', 'phone', 'desk', 'computer', 'mop_bucket', 'toaster', 'stovefan', 'dining_room', 'food_kiwi', 'wall', 'nightstand', 'oven', 'bed', 'doorjamb', 'mouse', 'dresser', 'cupboard', 'door', 'light', 'board_game', 'coffee']
Task:Pick up phone
response:
| {
"thoughts": "I walk to the kitchen and find the phone. I pick up the phone. I hold it to my ear.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <phone>",
"[FIND] <phone>",
"[GRAB] <phone>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['microwave', 'walllamp', 'bathroom', 'detergent', 'cpuscreen', 'ceiling', 'computer', 'toaster', 'brush', 'laptop', 'faucet', 'mousepad', 'clothes_pants', 'dirt', 'powersocket', 'oven', 'cup', 'bed', 'food_food', 'television', 'toothbrush_holder', 'tray', 'curtain', 'doorjamb', 'mouse', 'chessboard', 'laundry_detergent', 'sink', 'dishwasher', 'washing_machine']
Task:Wash clothes
response:
{
"thoughts": Walk to bathroom. Open washing maschine. Grab pants and put it into washing maschine. Pour laundry detergent into washing machine and close the washing maschine. Plug in washing machine.,
"actions": ['[WALK] <bathroom>', '[WALK] <washing_machine>', '[FIND] <washing_machine>', '[OPEN] <washing_machine>', '[FIND] <clothes_pants>', '[GRAB] <clothes_pants>', '[PUTBACK] <clothes_pants> <washing_machine>', '[WALK] <laundry_detergent>', '[FIND] <laundry_detergent>', '[GRAB] <laundry_detergent>', '[POUR] <laundry_detergent> <washing_machine>', '[CLOSE] <washing_machine>', '[PLUGIN] <washing_machine>']
}
|
Scene_Objects:['pot', 'food_pizza', 'walllamp', 'trashcan', 'knifeblock', 'drawing', 'needle', 'floor', 'ceiling', 'clothes_jacket', 'stovefan', 'toaster', 'window', 'homework', 'oil', 'dustpan', 'bowl', 'wall', 'pasta', 'oven', 'toy', 'food_food', 'oven_mitts', 'tray', 'kitchen_counter', 'napkin', 'table', 'dishwasher', 'laser_pointer', 'teeth']
Task:Cook some food
response:
| {
"thoughts": "I switch on the stove and cook my groceries there.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <oven>",
"[FIND] <oven>",
"[SWITCHON] <oven>",
"[FIND] <bowl>",
"[GRAB] <bowl>",
"[PUTBACK] <bowl> <oven>",
"[FIND] <food_food>",
"[SWITCHOFF] <oven>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_salt', 'freezer', 'walllamp', 'drawing', 'floor', 'ceiling', 'bookshelf', 'mail', 'table_cloth', 'shower', 'foundation', 'dustpan', 'microphone', 'wall', 'powersocket', 'food_food', 'check', 'television', 'rag', 'doorjamb', 'kitchen_counter', 'shoes', 'cupboard', 'door', 'sink', 'hairdryer', 'light', 'clothes_skirt', 'scrabble', 'creditcard']
Task:Turn off light
response:
{
"thoughts": Hit the switch.,
"actions": ['[WALK] <bedroom>', '[WALK] <light>', '[SWITCHOFF] <light>']
}
|
Scene_Objects:['glue', 'pot', 'blow_dryer', 'ceilingfan', 'drawing', 'bathroom_counter', 'floor', 'stovefan', 'ceilinglamp', 'balanceball', 'stereo', 'window', 'shower', 'clothes_socks', 'couch', 'candle', 'broom', 'wall', 'wall_clock', 'bed', 'television', 'curtain', 'pencil', 'kitchen_counter', 'dresser', 'napkin', 'spoon', 'table', 'dirt', 'creditcard']
Task:Relax on sofa
response:
| {
"thoughts": "Find the sofa and sit on it.",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[SIT] <couch>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['spectacles', 'scissors', 'food_ice_cream', 'chair', 'orchid', 'drawing', 'bills', 'knifeblock', 'floor', 'character', 'alcohol', 'mat', 'window', 'couch', 'microphone', 'pillow', 'dining_room', 'wall', 'pasta', 'clothes_gloves', 'check', 'coffe_maker', 'curtain', 'doorjamb', 'kitchen_counter', 'cupboard', 'table', 'dishwasher', 'colander', 'coffee']
Task:Set up table
response:
{
"thoughts": walk to entrance hall. find a table. pull table. walk to living rook. push table.,
"actions": ['[WALK] <home_office>', '[WALK] <table>', '[FIND] <table>', '[PULL] <table>', '[WALK] <home_office>', '[WALK] <table>', '[PUSH] <table>']
}
|
Scene_Objects:['tablelamp', 'food_salt', 'toilet', 'chair', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'desk', 'bedroom', 'keyboard', 'bookshelf', 'computer', 'mail', 'alcohol', 'toaster', 'ceilinglamp', 'balanceball', 'mousepad', 'food_orange', 'wall', 'nightstand', 'mouthwash', 'doorjamb', 'mouse', 'hairbrush', 'light']
Task:Write an email
response:
| {
"thoughts": "Turn on PC and log onto email account. Type email and send it. Switch off computer.",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[SWITCHON] <computer>",
"[FIND] <keyboard>",
"[TYPE] <keyboard>",
"[TURNTO] <computer>",
"[LOOKAT] <computer>",
"[SWITCHOFF] <computer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'juice', 'chair', 'freezer', 'walllamp', 'cpuscreen', 'food_egg', 'floor', 'ceiling', 'desk', 'bedroom', 'keyboard', 'computer', 'toaster', 'mat', 'ceilinglamp', 'faucet', 'picture', 'homework', 'mousepad', 'wall', 'oven', 'toy', 'newspaper', 'check', 'television', 'mouse', 'laundry_detergent', 'conditioner', 'laser_pointer']
Task:Push in desk chair
response:
{
"thoughts": Push the desk chair until it is under the desk.,
"actions": ['[WALK] <home_office>', '[WALK] <desk>', '[FIND] <chair>', '[TOUCH] <chair>', '[PUSH] <chair>']
}
|
Scene_Objects:['tablelamp', 'toilet', 'bathroom', 'ceilingfan', 'drawing', 'floor', 'ceiling', 'character', 'desk', 'tea_bag', 'remote_control', 'ceilinglamp', 'tvstand', 'balanceball', 'picture', 'window', 'couch', 'pillow', 'dining_room', 'wall', 'powersocket', 'cutting_board', 'novel', 'check', 'television', 'curtain', 'conditioner', 'hairdryer', 'clothes_dress', 'table']
Task:Watch TV
response:
| {
"thoughts": "watching TV from sofa",
"actions": [
"[WALK] <home_office>",
"[WALK] <television>",
"[FIND] <television>",
"[PLUGIN] <television>",
"[SWITCHON] <television>",
"[FIND] <couch>",
"[SIT] <couch>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[DROP] <remote_control>",
"[TURNTO] <television>",
"[LOOKAT] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'toilet', 'chair', 'bathroom', 'drawing', 'orchid', 'cpuscreen', 'floor', 'ceiling', 'desk', 'keyboard', 'computer', 'mat', 'ceilinglamp', 'tvstand', 'mousepad', 'closetdrawer', 'couch', 'pillow', 'wall', 'nightstand', 'novel', 'food_food', 'television', 'doorjamb', 'mouse', 'pencil', 'dresser', 'door', 'shoe_rack']
Task:Put in chair
response:
{
"thoughts": Take a comfortable chair. Place it in the entrance hall,
"actions": ['[WALK] <home_office>', '[WALK] <chair>', '[FIND] <chair>', '[GRAB] <chair>', '[WALK] <home_office>', '[WALK] <floor>', '[FIND] <floor>', '[PUTBACK] <chair> <floor>']
}
|
Scene_Objects:['dog', 'trashcan', 'drawing', 'bathroom_counter', 'floor', 'ceiling', 'phone', 'faucet', 'clothes_socks', 'microphone', 'wall', 'bench', 'powersocket', 'photoframe', 'food_rice', 'food_onion', 'towel_rack', 'cd', 'hairbrush', 'controller', 'maindoor', 'sink', 'bathroom_cabinet', 'home_office', 'clothes_skirt', 'vase', 'soap', 'instrument_violin', 'laser_pointer', 'hands_both']
Task:Wash hands
response:
| {
"thoughts": "I go to bathroom, go to sink, turn the water on via faucet, grab soap, use the soap to scrub my hands, put the soap back and wash my hands. Then I stop the water with faucet.",
"actions": [
"[WALK] <bathroom>",
"[WALK] <sink>",
"[FIND] <faucet>",
"[SWITCHON] <faucet>",
"[FIND] <soap>",
"[GRAB] <soap>",
"[FIND] <hands_both>",
"[SCRUB] <hands_both>",
"[PUTOBJBACK] <soap>",
"[WASH] <hands_both>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_pizza', 'filing_cabinet', 'crayon', 'walllamp', 'bathroom', 'toothbrush', 'floor', 'food_cake', 'toaster', 'plate', 'shower', 'faucet', 'electric_shaver', 'dish_soap', 'broom', 'wall', 'oven', 'bench', 'photoframe', 'light_bulb', 'novel', 'coffe_maker', 'television', 'tray', 'towel_rack', 'kettle', 'dishrack', 'table', 'dishwasher', 'tooth_paste']
Task:Wash dishes with dishwasher
response:
{
"thoughts": Go to kitchen to put dirty dishs into dishwasher with soap to clean the dirty dishes,
"actions": ['[WALK] <dining_room>', '[WALK] <dishwasher>', '[FIND] <dishwasher>', '[OPEN] <dishwasher>', '[FIND] <plate>', '[GRAB] <plate>', '[PUTBACK] <plate> <dishwasher>', '[FIND] <plate>', '[GRAB] <plate>', '[PUTBACK] <plate> <dishwasher>', '[FIND] <plate>', '[GRAB] <plate>', '[PUTBACK] <plate> <dishwasher>', '[FIND] <plate>', '[GRAB] <plate>', '[PUTBACK] <plate> <dishwasher>', '[FIND] <plate>', '[GRAB] <plate>', '[PUTBACK] <plate> <dishwasher>', '[FIND] <dish_soap>', '[GRAB] <dish_soap>', '[PUTBACK] <dish_soap> <dishwasher>', '[CLOSE] <dishwasher>', '[SWITCHON] <dishwasher>']
}
|
Scene_Objects:['filing_cabinet', 'scissors', 'jelly', 'walllamp', 'bathroom', 'ceilingfan', 'drawing', 'hanger', 'floor', 'ceiling', 'desk', 'remote_control', 'computer', 'ceilinglamp', 'tvstand', 'balanceball', 'shoe-shine_kit', 'window', 'couch', 'food_jam', 'wall', 'novel', 'television', 'curtain', 'doorjamb', 'comb', 'clothes_dress', 'sink', 'home_office', 'table']
Task:Change TV channel
response:
| {
"thoughts": "first i will go to living room and then take the remote and swtitch on television after that il change the channel",
"actions": [
"[WALK] <home_office>",
"[WALK] <television>",
"[FIND] <television>",
"[SWITCHON] <television>",
"[FIND] <remote_control>",
"[TURNTO] <remote_control>",
"[POINTAT] <remote_control>",
"[TOUCH] <remote_control>",
"[TURNTO] <television>",
"[WATCH] <television>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'microwave', 'chair', 'walllamp', 'orchid', 'floor', 'ceiling', 'computer', 'ceilinglamp', 'tvstand', 'window', 'closetdrawer', 'couch', 'microphone', 'form', 'pillow', 'wall', 'nightstand', 'oven', 'cup', 'coffe_maker', 'television', 'curtain', 'vacuum_cleaner', 'dishrack', 'dresser', 'hairbrush', 'door', 'table', 'soap']
Task:Watch TV
response:
{
"thoughts": I walk to the living room. I grab the remote control. I walk to the sofa. I sit down, and press the on button on the remote.,
"actions": ['[WALK] <home_office>', '[WALK] <remote_control>', '[FIND] <remote_control>', '[GRAB] <remote_control>', '[WALK] <couch>', '[SIT] <couch>', '[TOUCH] <remote_control>', '[FIND] <television>', '[SWITCHON] <television>']
}
|
Scene_Objects:['sheets', 'food_salt', 'walllamp', 'bathroom', 'trashcan', 'knifeblock', 'bedroom', 'toaster', 'ceilinglamp', 'tvstand', 'balanceball', 'shower', 'mousepad', 'microphone', 'dining_room', 'nightstand', 'basket_for_clothes', 'bed', 'newspaper', 'coffe_maker', 'check', 'tray', 'curtain', 'doorjamb', 'mouse', 'headset', 'laundry_detergent', 'maindoor', 'bathroom_cabinet', 'washing_machine']
Task:Wash clothes
response:
| {
"thoughts": "Walk to laundry room. Find laundry basket. Grab laundry basket. Find washing machine. Walk to washing machine. Open washing machine. Find detergent. Grab detergent. Put in detergent. Put in clothes. Close washing machine. Turn on machine.",
"actions": [
"[WALK] <bathroom>",
"[WALK] <basket_for_clothes>",
"[FIND] <basket_for_clothes>",
"[GRAB] <basket_for_clothes>",
"[FIND] <washing_machine>",
"[WALK] <washing_machine>",
"[OPEN] <washing_machine>",
"[FIND] <laundry_detergent>",
"[GRAB] <laundry_detergent>",
"[POUR] <laundry_detergent> <washing_machine>",
"[PUTOBJBACK] <laundry_detergent>",
"[PUTBACK] <basket_for_clothes> <washing_machine>",
"[CLOSE] <washing_machine>",
"[SWITCHON] <washing_machine>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sheets', 'tablelamp', 'microwave', 'toilet', 'chair', 'dog', 'walllamp', 'trashcan', 'cpuscreen', 'floor', 'ceiling', 'folder', 'desk', 'bookshelf', 'keyboard', 'thread', 'computer', 'mail', 'bookmark', 'mousepad', 'wall', 'nightstand', 'photoframe', 'novel', 'toilet_paper', 'tray', 'towel_rack', 'mouse', 'door', 'light']
Task:Read book
response:
{
"thoughts": Walk to the living room. Find a desk and grab a book. Walk to the chair, sit in the chair and read a book.,
"actions": ['[WALK] <home_office>', '[WALK] <desk>', '[FIND] <desk>', '[FIND] <novel>', '[GRAB] <novel>', '[WALK] <chair>', '[SIT] <chair>', '[READ] <novel>']
}
|
Scene_Objects:['filing_cabinet', 'food_ice_cream', 'food_carrot', 'longboard', 'ceiling', 'stamp', 'food_cake', 'stovefan', 'mat', 'oil', 'food_peanut_butter', 'dining_room', 'broom', 'pasta', 'basket_for_clothes', 'bed', 'coffe_maker', 'television', 'razor', 'tray', 'after_shave', 'mouse', 'laundry_detergent', 'clothes_scarf', 'sink', 'bathroom_cabinet', 'dishwasher', 'laser_pointer', 'clothes_shirt', 'washing_machine']
Task:Wash clothes
response:
| {
"thoughts": "Go to kitchen, find washing machine, find laundry, put laundry in washing machine, find soap, put soap in washing machine, turn on.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <washing_machine>",
"[FIND] <washing_machine>",
"[FIND] <basket_for_clothes>",
"[GRAB] <basket_for_clothes>",
"[OPEN] <washing_machine>",
"[PUTBACK] <basket_for_clothes> <washing_machine>",
"[WALK] <laundry_detergent>",
"[FIND] <laundry_detergent>",
"[GRAB] <laundry_detergent>",
"[POUR] <laundry_detergent> <washing_machine>",
"[CLOSE] <washing_machine>",
"[SWITCHON] <washing_machine>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['jelly', 'freezer', 'dog', 'chair', 'nail_polish', 'bathroom', 'orchid', 'bills', 'food_egg', 'hanger', 'character', 'bedroom', 'thread', 'toaster', 'faucet', 'mousepad', 'oil', 'nightstand', 'cup', 'photoframe', 'standingmirror', 'bed', 'video_game_controller', 'wine', 'doorjamb', 'maindoor', 'bathtub', 'vase', 'clothes_shirt', 'coffee']
Task:Read book
response:
{
"thoughts": pick up the book open the book and read,
"actions": ['[WALK] <bedroom>', '[WALK] <novel>', '[GRAB] <novel>', '[READ] <novel>']
}
|
Scene_Objects:['food_bread', 'filing_cabinet', 'mop', 'toilet', 'dog', 'freezer', 'longboard', 'floor', 'ceiling', 'purse', 'ceilinglamp', 'balanceball', 'faucet', 'foundation', 'food_jam', 'wall', 'oven', 'cup', 'bed', 'novel', 'food_food', 'coffe_maker', 'coffee_filter', 'curtain', 'console', 'pencil', 'maindoor', 'table', 'light', 'dirt']
Task:Put groceries in Fridge
response:
| {
"thoughts": "put away groceries in fridge. open the door, put on shelf, close the door.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <freezer>",
"[FIND] <freezer>",
"[OPEN] <freezer>",
"[FIND] <food_food>",
"[GRAB] <food_food>",
"[PUTIN] <food_food> <freezer>",
"[CLOSE] <freezer>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'juice', 'toilet', 'chair', 'freezer', 'dough', 'orchid', 'bathroom_counter', 'floor', 'ceiling', 'bookshelf', 'computer', 'toaster', 'mat', 'pillow', 'dining_room', 'wall', 'oven', 'photoframe', 'food_food', 'coffe_maker', 'tray', 'curtain', 'centerpiece', 'doorjamb', 'kitchen_counter', 'cupboard', 'laundry_detergent', 'bag', 'dishwasher']
Task:Put groceries in Fridge
response:
{
"thoughts": walk to kitchen, walk to fridge, look at bags, grab groceries, put groceries in fridge.,
"actions": ['[WALK] <dining_room>', '[WALK] <freezer>', '[FIND] <food_food>', '[GRAB] <food_food>', '[OPEN] <freezer>', '[PUTIN] <food_food> <freezer>']
}
|
Scene_Objects:['freezer', 'detergent', 'dough', 'drawing', 'bills', 'clothes_underwear', 'floor', 'bedroom', 'keyboard', 'piano_bench', 'ceilinglamp', 'balanceball', 'picture', 'window', 'foundation', 'electric_shaver', 'couch', 'dustpan', 'food_kiwi', 'wall', 'keys', 'check', 'television', 'tray', 'rag', 'shoes', 'door', 'table', 'vase', 'coffee']
Task:Relax on sofa
response:
| {
"thoughts": "Relax on the sofa in the living room.",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[SIT] <couch>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'food_bread', 'freezer', 'walllamp', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'bedroom', 'tea_bag', 'bookshelf', 'thread', 'alcohol', 'ceilinglamp', 'shoe-shine_kit', 'balanceball', 'wall', 'oven', 'food_food', 'coffe_maker', 'toilet_paper', 'check', 'food_rice', 'tray', 'towel_rack', 'pencil', 'kitchen_counter', 'napkin', 'dishwasher', 'clothes_shirt']
Task:Put groceries in Fridge
response:
{
"thoughts": Open the fridge, and put groceries in the fridge. Close the fridge.,
"actions": ['[WALK] <dining_room>', '[WALK] <freezer>', '[FIND] <freezer>', '[OPEN] <freezer>', '[FIND] <food_food>', '[GRAB] <food_food>', '[PUTIN] <food_food> <freezer>', '[CLOSE] <freezer>']
}
|
Scene_Objects:['juice', 'bills', 'hanger', 'shampoo', 'phone', 'dvd_player', 'character', 'alcohol', 'mat', 'stereo', 'clothes_socks', 'mousepad', 'wall', 'pasta', 'powersocket', 'tape', 'bed', 'novel', 'food_food', 'tray', 'coffee_pot', 'pencil', 'dishrack', 'cupboard', 'clothes_scarf', 'bathroom_cabinet', 'light', 'home_office', 'vase', 'scrabble']
Task:Pick up phone
response:
| {
"thoughts": "Walk into the home office. Pick up the phone.",
"actions": [
"[WALK] <home_office>",
"[WALK] <phone>",
"[FIND] <phone>",
"[GRAB] <phone>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'measuring_cup', 'toilet', 'chair', 'band-aids', 'freezer', 'walllamp', 'food_carrot', 'toothbrush', 'drawing', 'cpuscreen', 'bathroom_counter', 'floor', 'button', 'mat', 'tvstand', 'faucet', 'mousepad', 'food_kiwi', 'wall', 'light_bulb', 'coffe_maker', 'toothbrush_holder', 'rag', 'pencil', 'sink', 'bathroom_cabinet', 'soap', 'tooth_paste', 'teeth']
Task:Brush teeth
response:
{
"thoughts": getting ready for bed,
"actions": ['[WALK] <bathroom_counter>', '[FIND] <toothbrush>', '[GRAB] <toothbrush>', '[PUTBACK] <toothbrush> <bathroom_counter>', '[FIND] <tooth_paste>', '[SQUEEZE] <tooth_paste>', '[GRAB] <toothbrush>', '[SCRUB] <toothbrush>', '[RINSE] <toothbrush>', '[FIND] <toothbrush_holder>', '[PUTBACK] <toothbrush> <toothbrush_holder>']
}
|
Scene_Objects:['pot', 'filing_cabinet', 'food_salt', 'detergent', 'orchid', 'bathroom_counter', 'character', 'keyboard', 'remote_control', 'bookshelf', 'button', 'table_cloth', 'toaster', 'ceilinglamp', 'faucet', 'dustpan', 'nightstand', 'bench', 'tape', 'cup', 'bed', 'coffee_table', 'wooden_spoon', 'doorjamb', 'towel_rack', 'pencil', 'kitchen_counter', 'cd', 'dresser', 'hair']
Task:Change TV channel
response:
| {
"thoughts": "walk next to the coffee table. pick up the remote control from the top of the coffee table. Point the remote to the TV. Press the blue arrow on the remote.",
"actions": [
"[WALK] <home_office>",
"[WALK] <coffee_table>",
"[FIND] <remote_control>",
"[GRAB] <remote_control>",
"[FIND] <button>",
"[PUSH] <button>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['chair', 'walllamp', 'orchid', 'drawing', 'wallshelf', 'cpuscreen', 'floor', 'ceiling', 'bedroom', 'bookshelf', 'stovefan', 'mat', 'tvstand', 'ceilinglamp', 'shower', 'window', 'picture', 'mousepad', 'pillow', 'wall', 'nightstand', 'powersocket', 'photoframe', 'television', 'curtain', 'doorjamb', 'towel_rack', 'dresser', 'bathroom_cabinet', 'light']
Task:Watch TV
response:
{
"thoughts": Go to entrance hall, switch on TV, sit back in a chair and using remote watch your favorite program.,
"actions": ['[WALK] <home_office>', '[WALK] <television>', '[FIND] <television>', '[SWITCHON] <television>', '[WALK] <chair>', '[SIT] <chair>', '[TURNTO] <television>', '[WATCH] <television>', '[SWITCHOFF] <television>']
}
|
Scene_Objects:['pot', 'chair', 'walllamp', 'trashcan', 'knifeblock', 'floor', 'character', 'computer', 'table_cloth', 'toaster', 'plate', 'faucet', 'dish_soap', 'wall', 'oven', 'bench', 'cup', 'coffe_maker', 'food_food', 'television', 'wine', 'tray', 'pencil', 'kitchen_counter', 'dresser', 'cupboard', 'sink', 'home_office', 'dishwasher', 'vase']
Task:Wash dishes with dishwasher
response:
| {
"thoughts": "I walk into the kitchen and go to the dishwasher. I open the dishwasher and grab the dishes. Iplace the dishes into the dishwasher then close it. I add soap then turn the dishwasher on.",
"actions": [
"[WALK] <dining_room>",
"[WALK] <dishwasher>",
"[OPEN] <dishwasher>",
"[WALK] <kitchen_counter>",
"[FIND] <plate>",
"[GRAB] <plate>",
"[PUTBACK] <plate> <dishwasher>",
"[FIND] <dish_soap>",
"[GRAB] <dish_soap>",
"[PUTBACK] <dish_soap> <dishwasher>",
"[CLOSE] <dishwasher>",
"[SWITCHON] <dishwasher>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['filing_cabinet', 'tablelamp', 'juice', 'toilet', 'chair', 'jelly', 'cat', 'dough', 'cpuscreen', 'floor', 'desk', 'bedroom', 'keyboard', 'computer', 'laptop', 'mat', 'tvstand', 'mousepad', 'wall', 'nightstand', 'love_seat', 'cup', 'clothes_gloves', 'television', 'towel_rack', 'mouse', 'dresser', 'controller', 'bathroom_cabinet', 'electrical_outlet']
Task:Write book
response:
{
"thoughts": Writing a book,
"actions": ['[WALK] <home_office>', '[WALK] <laptop>', '[FIND] <laptop>', '[GRAB] <laptop>', '[FIND] <electrical_outlet>', '[PLUGIN] <laptop>', '[WALK] <love_seat>', '[SIT] <love_seat>', '[SWITCHON] <laptop>', '[FIND] <keyboard>', '[TYPE] <keyboard>']
}
|
Scene_Objects:['sheets', 'tablelamp', 'measuring_cup', 'dog', 'nail_polish', 'detergent', 'toothbrush', 'drawing', 'hanger', 'desk', 'computer', 'toaster', 'tvstand', 'iron', 'picture', 'window', 'shower', 'food_kiwi', 'bench', 'cup', 'food_cheese', 'photoframe', 'check', 'coffe_maker', 'television', 'kitchen_counter', 'cupboard', 'home_office', 'bathroom_cabinet', 'soap']
Task:Read book
response:
| {
"thoughts": "Read a book",
"actions": [
"[WALK] <home_office>",
"[WALK] <novel>",
"[FIND] <novel>",
"[GRAB] <novel>",
"[READ] <novel>",
"[PUTOBJBACK] <novel>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['pot', 'tablelamp', 'chair', 'walllamp', 'drawing', 'orchid', 'cpuscreen', 'wallshelf', 'hanger', 'floor', 'phone', 'ceiling', 'desk', 'keyboard', 'computer', 'mat', 'ceilinglamp', 'tvstand', 'picture', 'mousepad', 'couch', 'pillow', 'wall', 'nightstand', 'oven', 'powersocket', 'tape', 'check', 'television', 'tray', 'doorjamb', 'towel_rack', 'mouse', 'pencil', 'comb', 'dresser', 'door', 'light']
Task:Work
response:
{
"thoughts": Walk to home office, walk to desk, sit in chair, turn on computer, look at computer, click mouse and type on keyboard.,
"actions": ['[WALK] <home_office>', '[WALK] <desk>', '[FIND] <chair>', '[SIT] <chair>', '[FIND] <computer>', '[SWITCHON] <computer>', '[TURNTO] <computer>', '[LOOKAT] <computer>', '[FIND] <mouse>', '[TOUCH] <mouse>', '[FIND] <keyboard>', '[TYPE] <keyboard>', '[TURNTO] <computer>', '[LOOKAT] <computer>']
}
|
Scene_Objects:['food_ice_cream', 'beer', 'toilet', 'band-aids', 'detergent', 'ceilingfan', 'drawing', 'knifeblock', 'cpuscreen', 'hanger', 'desk', 'mail', 'computer', 'ceilinglamp', 'shoe-shine_kit', 'stereo', 'iron', 'shower', 'pillow', 'nightstand', 'toy', 'bed', 'food_food', 'oven_mitts', 'coffee_filter', 'after_shave', 'curtain', 'lighter', 'cupboard', 'maindoor']
Task:Pet cat
response:
| {
"thoughts": "I walk to the living room. I find the cat. I pet the cat.",
"actions": [
"[WALK] <home_office>",
"[WALK] <cat>",
"[FIND] <cat>",
"[TOUCH] <cat>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['food_pizza', 'tablelamp', 'juice', 'food_salt', 'measuring_cup', 'ceilingfan', 'orchid', 'shampoo', 'bedroom', 'desk', 'table_cloth', 'toaster', 'tvstand', 'stereo', 'plate', 'bench', 'basket_for_clothes', 'coffee_filter', 'ice', 'tray', 'pencil', 'headset', 'cupboard', 'maindoor', 'napkin', 'bag', 'light', 'soap', 'washing_machine', 'creditcard']
Task:Wash clothes
response:
{
"thoughts": Get dirty clothes out of basket for clothes in bedroom, carry dirty clothes to laundry room, place dirty clothes in washing machine, add laundry detergent to washing machine, turn washing machine on, set type of wash, start washing machine,
"actions": ['[WALK] <bedroom>', '[WALK] <basket_for_clothes>', '[FIND] <basket_for_clothes>', '[OPEN] <basket_for_clothes>', '[GRAB] <basket_for_clothes>', '[WALK] <bathroom>', '[WALK] <washing_machine>', '[FIND] <washing_machine>', '[OPEN] <washing_machine>', '[PUTBACK] <basket_for_clothes> <washing_machine>', '[FIND] <soap>', '[GRAB] <soap>', '[PUTBACK] <soap> <washing_machine>', '[FIND] <washing_machine>', '[CLOSE] <washing_machine>', '[SWITCHON] <washing_machine>']
}
|
Scene_Objects:['sheets', 'pot', 'tablelamp', 'freezer', 'walllamp', 'trashcan', 'food_butter', 'toothbrush', 'orchid', 'longboard', 'phone', 'keyboard', 'shower', 'stereo', 'oil', 'mousepad', 'clothes_pants', 'pillow', 'dining_room', 'nightstand', 'photoframe', 'newspaper', 'food_onion', 'mouse', 'shoes', 'cd', 'cupboard', 'controller', 'dishwasher', 'dirt']
Task:Turn on light
response:
| {
"thoughts": "I go home and it's dark so I can not see anything",
"actions": [
"[WALK] <home_office>",
"[WALK] <floor_lamp>",
"[FIND] <floor_lamp>",
"[SWITCHON] <floor_lamp>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['sheets', 'filing_cabinet', 'crayon', 'dog', 'freezer', 'knife', 'orchid', 'bathroom_counter', 'character', 'bedroom', 'faucet', 'couch', 'food_orange', 'towel', 'nightstand', 'oven', 'powersocket', 'photoframe', 'check', 'video_game_controller', 'food_donut', 'food_rice', 'food_onion', 'diary', 'dresser', 'hairbrush', 'clothes_scarf', 'napkin', 'bathroom_cabinet', 'colander']
Task:Pet cat
response:
{
"thoughts": Walk into the living room. Walk up to the cat. Pet the cat with your hand.,
"actions": ['[WALK] <home_office>', '[WALK] <cat>', '[FIND] <cat>', '[TOUCH] <cat>']
}
|
Scene_Objects:['tablelamp', 'microwave', 'bathroom', 'detergent', 'cat', 'drawing', 'floor', 'folder', 'keyboard', 'food_cake', 'table_cloth', 'ceilinglamp', 'deck_of_cards', 'window', 'couch', 'wall', 'wall_clock', 'powersocket', 'newspaper', 'television', 'pencil', 'maindoor', 'clothes_scarf', 'sink', 'home_office', 'table', 'tooth_paste', 'laser_pointer', 'teeth', 'creditcard']
Task:Pet cat
response:
| {
"thoughts": "walk to entrance hall, look at the sofa, sit on the sofa, look at the cat, allow cat to near with you, touch the cat smoothly on it head, now pet the cat smoothly",
"actions": [
"[WALK] <home_office>",
"[WALK] <couch>",
"[FIND] <couch>",
"[TURNTO] <couch>",
"[LOOKAT] <couch>",
"[SIT] <couch>",
"[FIND] <cat>",
"[TURNTO] <cat>",
"[LOOKAT] <cat>",
"[TOUCH] <cat>"
]
} |
You serve as an AI task planner.
1. Your task is to create a plan to achieve a goal by converting it into a sequence of actions. Each action should follow the format "[verb] <target1> <target2>", where 'verb' represents the action, and 'target1' and 'target2' are optional arguments.
2. You must describe your plan in natural language at the beginning. After that, you should list all the actions together.
Ensure that you response can be parsed by Python json.loads.
Here is an example plan to achieve a goal for reference:
Scene_Objects:['tablelamp', 'chair', 'cloth_napkin', 'walllamp', 'orchid', 'cpuscreen', 'wallshelf', 'hanger', 'floor', 'ceiling', 'phone', 'desk', 'keyboard', 'computer', 'ceilinglamp', 'tvstand', 'mat', 'mousepad', 'couch', 'pillow', 'wall', 'nightstand', 'television', 'doorjamb', 'mouse', 'pencil', 'headset', 'dresser', 'door', 'sink']
Task:Clean
response:
{
"thoughts": Use a cloth to wipe away any marks on the computer screen.,
"actions": ['[WALK] <home_office>', '[WALK] <computer>', '[FIND] <cloth_napkin>', '[GRAB] <cloth_napkin>', '[WIPE] <computer>']
}
|
Scene_Objects:['filing_cabinet', 'tablelamp', 'chair', 'trashcan', 'ceilingfan', 'drawing', 'cpuscreen', 'floor', 'ceiling', 'phone', 'character', 'desk', 'keyboard', 'bookshelf', 'envelope', 'computer', 'mail', 'ceilinglamp', 'balanceball', 'mousepad', 'wall', 'nightstand', 'check', 'food_food', 'television', 'tray', 'doorjamb', 'mouse', 'chessboard', 'light']
Task:Browse internet
response:
| {
"thoughts": "Walk to home office. Walk to computer. Switch on Computer.",
"actions": [
"[WALK] <home_office>",
"[WALK] <computer>",
"[SWITCHON] <computer>"
]
} |
Subsets and Splits