so101_grab_mofang_v1

该数据集使用LeRobot创建，专注于机器人操作任务，涉及so_follower类型的机器人。数据集结构包括动作数据（如肩部、肘部、腕部和抓取器的位置）、观测状态（与动作相同的关节位置）、来自正面和侧面的视频观测（分辨率为480x640，帧率为30fps，采用AV1编码），以及元数据（如时间戳、帧索引、episode索引等）。总共有50个episodes，包含17324帧数据，数据以parquet文件格式存储，视频以mp4格式存储，主要用于训练目的。This dataset was developed using LeRobot, focusing on robotic manipulation tasks involving so_follower-type robots. The dataset structure encompasses action data (e.g., positions of the shoulder, elbow, wrist, and gripper), observation states consistent with the joint positions of the actions, video observations from front and side viewpoints with a resolution of 480x640, a frame rate of 30fps and AV1 encoding, as well as metadata including timestamps, frame indices, episode indices and others. In total, there are 50 episodes containing 17,324 frames of data. The tabular data is stored in Parquet file format, while the videos are stored in MP4 format, and this dataset is primarily intended for training purposes.