stanford_kuka_multimodal_dataset_converted_externally_to_rlds

  • Keterangan :

Penyisipan pasak Kuka iiwa dengan umpan balik paksa

Membelah Contoh
'train' 3.000
  • Struktur fitur :
FeaturesDict({
    'episode_metadata': FeaturesDict({
    }),
    'steps': Dataset({
        'action': Tensor(shape=(4,), dtype=float32, description=Robot action, consists of [3x EEF position, 1x gripper open/close].),
        'discount': Scalar(shape=(), dtype=float32, description=Discount if provided, default to 1.),
        '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({
            'contact': Tensor(shape=(50,), dtype=float32, description=Robot contact information.),
            'depth_image': Tensor(shape=(128, 128, 1), dtype=float32, description=Main depth camera observation.),
            'ee_forces_continuous': Tensor(shape=(50, 6), dtype=float32, description=Robot end-effector forces.),
            'ee_orientation': Tensor(shape=(4,), dtype=float32, description=Robot end-effector orientation quaternion.),
            'ee_orientation_vel': Tensor(shape=(3,), dtype=float32, description=Robot end-effector orientation velocity.),
            'ee_position': Tensor(shape=(3,), dtype=float32, description=Robot end-effector position.),
            'ee_vel': Tensor(shape=(3,), dtype=float32, description=Robot end-effector velocity.),
            'ee_yaw': Tensor(shape=(4,), dtype=float32, description=Robot end-effector yaw.),
            'ee_yaw_delta': Tensor(shape=(4,), dtype=float32, description=Robot end-effector yaw delta.),
            'image': Image(shape=(128, 128, 3), dtype=uint8, description=Main camera RGB observation.),
            'joint_pos': Tensor(shape=(7,), dtype=float32, description=Robot joint positions.),
            'joint_vel': Tensor(shape=(7,), dtype=float32, description=Robot joint velocities.),
            'optical_flow': Tensor(shape=(128, 128, 2), dtype=float32, description=Optical flow.),
            'state': Tensor(shape=(8,), dtype=float32, description=Robot proprioceptive information, [7x joint pos, 1x gripper open/close].),
        }),
        'reward': Scalar(shape=(), dtype=float32, description=Reward if provided, 1 on final step for demos.),
    }),
})
  • Dokumentasi fitur :
Fitur Kelas Membentuk Tipe D Keterangan
FiturDict
episode_metadata FiturDict
tangga Kumpulan data
langkah/tindakan Tensor (4,) float32 Aksi robot, terdiri dari [3x posisi EEF, 1x gripper buka/tutup].
langkah/diskon Skalar float32 Diskon jika disediakan, defaultnya adalah 1.
langkah/adalah_pertama Tensor bodoh
langkah/adalah_terakhir Tensor bodoh
langkah/is_terminal Tensor bodoh
langkah/bahasa_penyematan Tensor (512,) float32 Penyematan bahasa Kona. Lihat https://tfhub.dev/google/universal-sentence-encoder-large/5
langkah/bahasa_instruksi Teks rangkaian Instruksi Bahasa.
langkah/pengamatan FiturDict
langkah/observasi/kontak Tensor (50,) float32 Informasi kontak robot.
langkah/pengamatan/kedalaman_gambar Tensor (128, 128, 1) float32 Pengamatan kamera kedalaman utama.
langkah/pengamatan/ee_forces_continuous Tensor (50, 6) float32 Kekuatan efektor akhir robot.
langkah/pengamatan/ee_orientation Tensor (4,) float32 Angka empat orientasi efektor akhir robot.
langkah/pengamatan/ee_orientation_vel Tensor (3,) float32 Kecepatan orientasi efektor akhir robot.
langkah/pengamatan/ee_position Tensor (3,) float32 Posisi efektor akhir robot.
langkah/pengamatan/ee_vel Tensor (3,) float32 Kecepatan efektor akhir robot.
langkah/pengamatan/ee_yaw Tensor (4,) float32 Efektor akhir robot menguap.
langkah/pengamatan/ee_yaw_delta Tensor (4,) float32 Delta yaw efektor akhir robot.
langkah/pengamatan/gambar Gambar (128, 128, 3) uint8 Pengamatan RGB kamera utama.
langkah/pengamatan/joint_pos Tensor (7,) float32 Posisi sendi robot.
langkah/pengamatan/joint_vel Tensor (7,) float32 Kecepatan sambungan robot.
langkah/pengamatan/optical_flow Tensor (128, 128, 2) float32 Aliran optik.
langkah/pengamatan/keadaan Tensor (8,) float32 Informasi proprioseptif robot, [7x joint pos, 1x gripper buka/tutup].
langkah/hadiah Skalar float32 Hadiah jika diberikan, 1 pada langkah terakhir untuk demo.
@inproceedings{lee2019icra,
  title={Making sense of vision and touch: Self-supervised learning of multimodal representations for contact-rich tasks},
  author={Lee, Michelle A and Zhu, Yuke and Srinivasan, Krishnan and Shah, Parth and Savarese, Silvio and Fei-Fei, Li and  Garg, Animesh and Bohg, Jeannette},
  booktitle={2019 IEEE International Conference on Robotics and Automation (ICRA)},
  year={2019},
  url={https://arxiv.org/abs/1810.10191}
}