blue_cube_in_mug_20260620_122606

该数据集是一个机器人学习数据集，使用LeRobot工具创建，专门用于训练机器人执行任务。它包含了机器人执行任务时的动作和观察数据，具体特征包括：6维浮点型动作向量（对应肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置）、6维浮点型观察状态向量（与动作相同）、前摄像头视频观察（分辨率为480x640，3通道，30帧/秒，AV1编码）、时间戳、帧索引、剧集索引、索引和任务索引。数据集总共有7个剧集，3198帧，1个任务，数据文件大小为100MB，视频文件大小为200MB，机器人类型为so_follower，所有数据均用于训练。This dataset is a robot learning dataset created using the LeRobot tool, specifically designed for training robots to perform tasks. It includes action and observation data during task execution, with features such as: 6-dimensional floating-point action vectors (corresponding to shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position), 6-dimensional floating-point observation state vectors (same as actions), front camera video observations (resolution 480x640, 3 channels, 30 fps, AV1 encoding), timestamps, frame indices, episode indices, indices, and task indices. The dataset consists of 7 episodes, 3198 frames, 1 task, with a data file size of 100MB and video file size of 200MB. The robot type is so_follower, and all data is used for training.