YOLO2431/stack_cup_v0.1

该数据集由LeRobot创建，是一个用于机器人学研究的双手机器人数据集，包含雅马哈双手机器人（yam_bimanual）的演示数据。数据集总计105个episodes，166070帧，帧率为30fps，涵盖2个不同任务，并包含315个关联视频。数据以parquet格式存储，视频为mp4格式。特征包括观察状态（32维浮点数组，编码左右机械臂的3D位置、6D旋转表示、夹爪开合状态和6个关节角度）、动作（与观察状态同构）、三个相机视角的图像观察（头部、左手腕和右手腕，均为720x1280分辨率RGB视频），以及时间戳、帧索引、episode索引、索引和任务索引等元数据。数据集适用于机器人模仿学习、强化学习和行为克隆等任务。This dataset was created using LeRobot and is a bimanual robotics dataset for research, containing demonstration data from a Yamaha bimanual robot (yam_bimanual). It includes 105 episodes, 166070 frames at 30fps, covering 2 distinct tasks, with 315 associated videos. Data is stored in parquet format, and videos are in mp4 format. Features include observation state (a 32-dimensional float32 array encoding 3D positions, 6D rotation representations, gripper states, and 6 joint angles for both left and right arms), actions (isomorphic to observation state), image observations from three camera views (head, left wrist, and right wrist, all RGB videos at 720x1280 resolution), and metadata such as timestamp, frame index, episode index, index, and task index. The dataset is suitable for tasks like robot imitation learning, reinforcement learning, and behavior cloning.