grab_the_duck2

该数据集是一个机器人学数据集，使用LeRobot创建，专门用于机器人控制任务。数据集包含45个episodes，总计38744帧，涉及一个任务。机器人类型为so_follower。数据以parquet文件格式存储，包括动作特征（如肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置）、观测状态（关节位置）、前视摄像头视频（分辨率1080x1920，30fps）和腕部摄像头视频（分辨率480x640，30fps）。数据集还包含时间戳、帧索引、episode索引等元数据，适用于机器人学习和控制研究。This dataset is a robotics dataset created using LeRobot, specifically designed for robot control tasks. It contains 45 episodes, totaling 38,744 frames, and involves one task. The robot type is so_follower. Data is stored in parquet format, including action features (such as shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position), observation states (joint positions), front-view camera video (resolution 1080x1920, 30fps), and wrist camera video (resolution 480x640, 30fps). The dataset also includes metadata such as timestamps, frame indices, and episode indices, making it suitable for robotics learning and control research.