berkeley_mvp_converted_externally_to_rlds

  • 説明

6 つの操作タスクを実行する xArm

スプリット
'train' 480
  • 機能の構造:
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({
            'gripper': Scalar(shape=(), dtype=bool),
            'hand_image': Image(shape=(480, 640, 3), dtype=uint8),
            'joint_pos': Tensor(shape=(7,), dtype=float32),
            'pose': Tensor(shape=(7,), dtype=float32),
        }),
        'reward': Scalar(shape=(), dtype=float32),
    }),
})
  • 機能ドキュメント:
特徴クラスDタイプ説明
特徴辞書
エピソード_メタデータ特徴辞書
エピソードメタデータ/ファイルパス文章元のデータ ファイルへのパス。
ステップデータセット
ステップ/アクションテンソル(8,) float32ロボットのアクションは、[デルタ関節 7 点、グリッパー 2 値ステート 1 つ] で構成されます。
歩数/割引スカラーfloat32割引が指定されている場合、デフォルトは 1 です。
ステップ/is_firstテンソルブール
ステップ/is_lastテンソルブール
ステップ/is_terminalテンソルブール
ステップ/言語_埋め込みテンソル(512,) float32コナ言語の埋め込み。 https://tfhub.dev/google/universal-sentence-encoder-large/5を参照してください。
ステップ/言語説明文章言語指導。
ステップ/観察特徴辞書
ステップ/観察/グリッパースカラーブールバイナリ グリッパ状態 (1 - 閉じ、0 - 開き)
歩数/観察/手の画像画像(480、640、3) uint8ハンドカメラRGB観察。
歩数/観察/joint_posテンソル(7、) float32 xArm ジョイントの位置 (7 DoF)。
ステップ/観察/ポーズテンソル(7、) float32グリッパーポーズ、ロボットフレーム、[3位置4回転]
歩数/報酬スカラーfloat32提供されている場合は報酬、デモの最終ステップで 1。
  • 引用
@InProceedings{Radosavovic2022,
  title = {Real-World Robot Learning with Masked Visual Pre-training},
  author = {Ilija Radosavovic and Tete Xiao and Stephen James and Pieter Abbeel and Jitendra Malik and Trevor Darrell},
  booktitle = {CoRL},
  year = {2022}
}