robot-learning/ex1_all_v2

该数据集使用LeRobot创建，是一个机器人学习数据集，专门用于机器人控制任务。数据集包含168个episodes，总计104906个frames，涵盖2个不同的任务。数据以parquet文件格式存储，视频文件以MP4格式提供，帧率为30fps。数据集特征包括：动作（action），表示6维关节位置（如肩部旋转、肩部提升、肘部弯曲、手腕弯曲、手腕旋转和夹爪位置）；观测状态（observation.state），同样为6维关节位置；前摄像头图像（observation.images.front），分辨率为480x640，3通道彩色视频；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强化学习、模仿学习等研究，机器人类型为so_follower。This dataset was created using LeRobot and is a robotics learning dataset designed for robot control tasks. It contains 168 episodes, totaling 104906 frames, and covers 2 distinct tasks. The data is stored in parquet format, with video files in MP4 format at 30fps. The dataset features include: action, representing 6-dimensional joint positions (e.g., shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper positions); observation state, also 6-dimensional joint positions; front camera images (observation.images.front) with a resolution of 480x640 and 3 color channels; and metadata such as timestamp, frame index, episode index, index, and task index. The dataset is suitable for research in robotics reinforcement learning, imitation learning, etc., with the robot type being so_follower.