asu_tabletop_converted_externally_to_rlds

  • תיאור :

UR5 מבצע משימות בחירה/מקום/סיבוב בראש השולחן

לְפַצֵל דוגמאות
'train' 110
  • מבנה תכונה :
FeaturesDict({
   
'episode_metadata': FeaturesDict({
       
'file_path': Text(shape=(), dtype=string),
   
}),
   
'steps': Dataset({
       
'action': Tensor(shape=(7,), dtype=float32, description=Robot action, consists of [7x joint velocities, 2x gripper velocities, 1x terminate episode].),
       
'action_delta': Tensor(shape=(7,), dtype=float32, description=Robot delta action, consists of [7x joint velocities, 2x gripper velocities, 1x terminate episode].),
       
'action_inst': Text(shape=(), dtype=string),
       
'discount': Scalar(shape=(), dtype=float32, description=Discount if provided, default to 1.),
       
'goal_object': Text(shape=(), dtype=string),
       
'ground_truth_states': FeaturesDict({
           
'EE': Tensor(shape=(6,), dtype=float32, description=xyzrpy),
           
'bottle': Tensor(shape=(6,), dtype=float32, description=xyzrpy),
           
'bread': Tensor(shape=(6,), dtype=float32, description=xyzrpy),
           
'coke': Tensor(shape=(6,), dtype=float32, description=xyzrpy),
           
'cube': Tensor(shape=(6,), dtype=float32, description=xyzrpy),
           
'milk': Tensor(shape=(6,), dtype=float32, description=xyzrpy),
           
'pepsi': Tensor(shape=(6,), dtype=float32, description=xyzrpy),
       
}),
       
'is_first': bool,
       
'is_last': bool,
       
'is_terminal': bool,
       
'language_embedding': Tensor(shape=(512,), dtype=float32, description=Kona language embedding. See https://tfhub.dev/google/universal-sentence-encoder-large/5),
       
'language_instruction': Text(shape=(), dtype=string),
       
'observation': FeaturesDict({
           
'image': Image(shape=(224, 224, 3), dtype=uint8, description=Main camera RGB observation.),
           
'state': Tensor(shape=(7,), dtype=float32, description=Robot state, consists of [6x robot joint angles, 1x gripper position].),
           
'state_vel': Tensor(shape=(7,), dtype=float32, description=Robot joint velocity, consists of [6x robot joint angles, 1x gripper position].),
       
}),
       
'reward': Scalar(shape=(), dtype=float32, description=Reward if provided, 1 on final step for demos.),
   
}),
})
  • תיעוד תכונה :
תכונה מַחלָקָה צוּרָה Dtype תֵאוּר
FeaturesDict
episode_metadata FeaturesDict
episode_metadata/file_path טֶקסט חוּט נתיב לקובץ הנתונים המקורי.
צעדים מערך נתונים
צעדים/פעולה מוֹתֵחַ (7,) לצוף32 פעולת רובוט, מורכבת מ[7x מהירויות מפרקים, 2x מהירויות תפסן, 1x פרק סיום].
steps/action_delta מוֹתֵחַ (7,) לצוף32 פעולת דלתא של רובוט, מורכבת מ[7x מהירויות מפרקים, 2x מהירויות תפסן, 1x פרק סיום].
steps/action_inst טֶקסט חוּט פעולה שיש לבצע.
צעדים/הנחה סקלר לצוף32 הנחה אם ניתנת, ברירת המחדל היא 1.
צעדים/אובייקט_מטרה טֶקסט חוּט אובייקט שיש לתמרן איתו.
צעדים/מצבי_אמת_קרקע FeaturesDict
steps/ground_truth_states/EE מוֹתֵחַ (6,) לצוף32 xyzrpy
צעדים/מצב_אמת_קרקע/בקבוק מוֹתֵחַ (6,) לצוף32 xyzrpy
צעדים/מצבי_אמת_קרקע/לחם מוֹתֵחַ (6,) לצוף32 xyzrpy
צעדים/מצבי_אמת_קרקע/קולה מוֹתֵחַ (6,) לצוף32 xyzrpy
steps/ground_truth_states/cube מוֹתֵחַ (6,) לצוף32 xyzrpy
צעדים/מצבי_אמת_קרקע/חלב מוֹתֵחַ (6,) לצוף32 xyzrpy
steps/ground_truth_states/pepsi מוֹתֵחַ (6,) לצוף32 xyzrpy
צעדים/הוא_ראשון מוֹתֵחַ bool
צעדים/הוא_אחרון מוֹתֵחַ bool
steps/is_terminal מוֹתֵחַ bool
שלבים/הטבעת_שפה מוֹתֵחַ (512,) לצוף32 הטבעת שפת Kona. ראה https://tfhub.dev/google/universal-sentence-encoder-large/5
שלבים/הוראת_שפה טֶקסט חוּט הוראת שפה.
צעדים/תצפית FeaturesDict
צעדים/תצפית/תמונה תְמוּנָה (224, 224, 3) uint8 תצפית RGB של מצלמה ראשית.
צעדים/תצפית/מצב מוֹתֵחַ (7,) לצוף32 מצב רובוט, מורכב מ[6x זוויות מפרק רובוט, 1x מיקום תפס].
צעדים/תצפית/מצב_vel מוֹתֵחַ (7,) לצוף32 מהירות מפרק הרובוט, מורכבת מ[6x זוויות מפרק רובוט, 1x מיקום תפס].
צעדים/פרס סקלר לצוף32 תגמול אם מסופק, 1 בשלב האחרון להדגמות.
  • ציטוט :
@inproceedings{zhou2023modularity,
  title
={Modularity through Attention: Efficient Training and Transfer of Language-Conditioned Policies for Robot Manipulation},
  author
={Zhou, Yifan and Sonawani, Shubham and Phielipp, Mariano and Stepputtis, Simon and Amor, Heni},
  booktitle
={Conference on Robot Learning},
  pages
={1684--1695},
  year
={2023},
  organization
={PMLR}
}
@article{zhou2023learning,
  title
={Learning modular language-conditioned robot policies through attention},
  author
={Zhou, Yifan and Sonawani, Shubham and Phielipp, Mariano and Ben Amor, Heni and Stepputtis, Simon},
  journal
={Autonomous Robots},
  pages
={1--21},
  year
={2023},
  publisher
={Springer}
}