blue_marker

该数据集是一个使用LeRobot创建的机器人数据集，包含51个episodes，总计11692个frames，数据以parquet格式存储。特征包括动作（action），为6维浮点数组，表示机器人关节位置（肩部平移、肩部升降、肘部弯曲、腕部弯曲、腕部旋转、夹爪位置）；观测状态（observation.state），同样为6维浮点数组，表示关节位置；观测图像（observation.images.overhead和observation.images.wrist），均为视频数据，分辨率480x640，3通道，帧率30fps，使用av1编解码器；以及时间戳、帧索引、episode索引、索引和任务索引等元数据。机器人类型为so_follower。数据集总大小：数据文件100MB，视频文件200MB。数据分割为训练集（全部51个episodes）。This dataset is a robotics dataset created using LeRobot. It contains 51 episodes, totaling 11692 frames, and is stored in parquet format. Features include: action, a 6-dimensional float array representing robot joint positions (shoulder translation, shoulder lift, elbow bend, wrist bend, wrist rotation, gripper position); observation.state, also a 6-dimensional float array for joint positions; observation images (observation.images.overhead and observation.images.wrist), both video data with resolution 480x640, 3 channels, frame rate 30fps, using the av1 codec; and metadata such as timestamp, frame index, episode index, index, and task index. The robot type is so_follower. Total dataset size: data file 100MB, video files 200MB. Data split: training set (all 51 episodes).