dish_pick_phaseC50_20260529_232425

该数据集使用LeRobot工具创建，专注于机器人技术领域，包含50个episodes，总计28624帧数据，帧率为30fps。数据特征包括：机器人动作（6个关节位置：肩部平移、肩部升降、肘部弯曲、腕部弯曲、腕部旋转、夹爪位置）、观测状态（与动作相同的6个关节位置）、图像观测（前视和腕部摄像头视频，分辨率480x640，RGB三通道，AV1编码）、时间戳、帧索引、episode索引、任务索引等。数据以parquet格式存储，视频以MP4格式存储，机器人类型为so_follower。数据集用于机器人控制与视觉感知的机器学习任务，如模仿学习或强化学习。This dataset is created using the LeRobot tool, focusing on the field of robotics. It contains 50 episodes with a total of 28,624 frames at a frame rate of 30 fps. The data features include robot actions (6 joint positions: shoulder translation, shoulder lift, elbow flexion, wrist flexion, wrist rotation, gripper position), observation states (the same 6 joint positions as the robot actions), image observations (videos from the front-facing and wrist-mounted cameras with a resolution of 480×640, 3-channel RGB and AV1 encoding), timestamps, frame indices, episode indices, task indices, and more. The data is stored in Parquet format while the videos are stored in MP4 format, and the robot type is so_follower. This dataset is intended for machine learning tasks related to robot control and visual perception, such as imitation learning or reinforcement learning.