cmu_stretch

  • 설명 :

안녕하세요 스트레치 로봇 주방 상호작용

나뉘다
'train' 135
  • 기능 구조 :
FeaturesDict({
    'episode_metadata': FeaturesDict({
        'file_path': Text(shape=(), dtype=string),
    }),
    'steps': Dataset({
        'action': Tensor(shape=(8,), dtype=float32),
        'discount': Scalar(shape=(), dtype=float32),
        'is_first': bool,
        'is_last': bool,
        'is_terminal': bool,
        'language_embedding': Tensor(shape=(512,), dtype=float32),
        'language_instruction': Text(shape=(), dtype=string),
        'observation': FeaturesDict({
            'image': Image(shape=(128, 128, 3), dtype=uint8),
            'state': Tensor(shape=(4,), dtype=float32),
        }),
        'reward': Scalar(shape=(), dtype=float32),
    }),
})
  • 기능 문서 :
특징 수업 모양 Dtype 설명
특징Dict
에피소드_메타데이터 특징Dict
에피소드_메타데이터/파일_경로 텍스트 원본 데이터 파일의 경로입니다.
단계 데이터세트
단계/작업 텐서 (8,) float32 로봇 액션은 [3x ee pos, 3x ee rot, 1x 그리퍼 바이너리 액션, 1x 종료 에피소드]로 구성됩니다.
걸음수/할인 스칼라 float32 할인이 제공되면 기본값은 1입니다.
걸음수/is_first 텐서 부울
걸음수/is_last 텐서 부울
단계/is_terminal 텐서 부울
단계/언어_임베딩 텐서 (512,) float32 코나 언어 임베딩. https://tfhub.dev/google/universal-sentence-encoder-large/5를 참조하세요.
단계/언어_지시 텍스트 언어 교육.
단계/관찰 특징Dict
단계/관찰/이미지 영상 (128, 128, 3) uint8 메인 카메라 RGB 관찰.
단계/관찰/상태 텐서 (4,) float32 로봇 상태는 [3x 로봇 관절 각도/ee 위치, 1x 그리퍼 위치]로 구성됩니다.
걸음수/보상 스칼라 float32 제공되는 경우 보상, 데모의 마지막 단계에서 1개.
  • 인용 :
@inproceedings{bahl2023affordances,
  title={Affordances from Human Videos as a Versatile Representation for Robotics},
  author={Bahl, Shikhar and Mendonca, Russell and Chen, Lili and Jain, Unnat and Pathak, Deepak},
  booktitle={CVPR},
  year={2023}
}
@article{mendonca2023structured,
  title={Structured World Models from Human Videos},
  author={Mendonca, Russell and Bahl, Shikhar and Pathak, Deepak},
  journal={CoRL},
  year={2023}
}