witsense-ai/so101_tidy-up-table
收藏资源简介:
该数据集使用LeRobot创建,是一个机器人学数据集,专注于机器人控制任务。数据集包含17个训练episodes,总计15148帧,涉及1个具体任务(例如整理桌子)。数据以parquet文件格式存储,视频文件为mp4格式。特征包括: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)、状态观察(与动作相同的6个关节位置)、顶部摄像头和腕部摄像头的图像观察(均为480x640分辨率、3通道RGB视频,帧率30fps),以及时间戳、帧索引、episode索引等元数据。机器人类型为so_follower,适用于机器人模仿学习或强化学习场景。
This dataset was created using LeRobot and is a robotics dataset focused on robot control tasks. It contains 17 training episodes, totaling 15148 frames, and involves 1 specific task (e.g., tidying up a table). The data is stored in parquet format, with video files in mp4 format. Features include: action (6 joint positions: shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, gripper position), observation.state (same 6 joint positions as action), observation images from top and wrist cameras (both 480x640 resolution, 3-channel RGB video at 30fps), and metadata such as timestamp, frame index, episode index, etc. The robot type is so_follower, suitable for robot imitation learning or reinforcement learning scenarios.




