multitask-so101-50eps-per-task

该数据集是一个机器人任务数据集，由LeRobot创建，专为so_follower机器人设计。它包含154个episodes，总计117,445帧，覆盖4个不同任务。数据以parquet文件格式存储，视频文件为mp4格式，帧率为30fps。数据集特征包括动作（6维浮点数组，控制肩部、肘部、腕部和夹持器的位置）、观察状态（6维浮点数组，表示机器人关节位置）、顶部图像（480x640x3的视频，提供机器人顶部视角）和手腕图像（480x640x3的视频，提供机器人手腕视角）。此外，还包括时间戳、帧索引、episode索引、索引和任务索引等元数据。该数据集适用于机器人学习、控制任务和视觉导航等研究领域。

This dataset is a robotic task dataset created by LeRobot, specifically designed for the so_follower robot. It contains 154 episodes, totaling 117,445 frames, covering 4 different tasks. The data is stored in parquet file format, with video files in mp4 format at a frame rate of 30fps. The dataset features include actions (6-dimensional floating-point arrays controlling the positions of the shoulder, elbow, wrist, and gripper), observation states (6-dimensional floating-point arrays representing robot joint positions), top images (480x640x3 videos providing a top-down view of the robot), and wrist images (480x640x3 videos providing a wrist perspective of the robot). Additionally, it includes metadata such as timestamps, frame indices, episode indices, indices, and task indices. The dataset is suitable for research areas such as robotic learning, control tasks, and visual navigation.