klvarshaa/so101_arrange_table_50ep_v2
收藏资源简介:
这是一个使用LeRobot创建的机器人任务数据集,专门用于机器人学习与控制研究。数据集包含50个训练episodes,总计79,486帧数据,任务类型为so_follower。数据特征包括:机器人动作(6个关节位置:肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转、夹爪位置)、状态观测(与动作相同的6个关节位置)、以及三个视角的视觉观测(顶部、侧面和腕部摄像头拍摄的480x640分辨率RGB视频,帧率30fps)。此外,数据集还包含时间戳、帧索引、episode索引、任务索引等元数据。数据以parquet格式存储,视频使用AV1编码。
This is a robotic task dataset developed with LeRobot, exclusively tailored for robotic learning and control research. The dataset comprises 50 training episodes, totaling 79,486 frames, with the task type being so_follower. The data features include: robotic actions (6 joint positions: shoulder translation, shoulder lift, elbow flexion, wrist flexion, wrist rotation, and gripper position), state observations (the same 6 joint positions as the actions), and three-view visual observations: 480×640 resolution RGB videos captured by top, side, and wrist cameras, with a frame rate of 30 fps. In addition, the dataset contains metadata such as timestamps, frame indices, episode indices, and task indices. The data is stored in Parquet format, and the videos are encoded using AV1.




