lyl472324464/2026-05-05_direction-water-with-rinse

该数据集是一个用于机器人控制的数据集，由ALOHA双机械臂系统采集。它包含132个episodes，总计183430帧，采样频率为50Hz。数据特征包括机器人的状态（14维关节位置）、动作（14维关节控制）、速度（14维关节速度）和力矩（14维关节力矩），以及来自四个摄像头的视频数据：高位摄像头、低位摄像头、左手腕摄像头和右手腕摄像头，每个视频的分辨率为480x640，格式为MP4（AV1编码）。数据集还包括任务索引、子任务标签、训练标志、时间戳、帧索引、episode索引等元数据。数据以parquet文件存储，视频文件单独存储，总数据量约为400MB（数据和视频各200MB）。该数据集适用于机器人学习、模仿学习或强化学习任务。数据集基于LeRobot项目创建，使用Apache 2.0许可证。This dataset is a robotics control dataset collected from an ALOHA dual-arm robot system. It contains 132 episodes with a total of 183,430 frames at a sampling rate of 50Hz. The features include robot state (14-dimensional joint positions), actions (14-dimensional joint controls), velocity (14-dimensional joint velocities), and effort (14-dimensional joint efforts), along with video data from four cameras: high camera, low camera, left wrist camera, and right wrist camera, each with a resolution of 480x640 in MP4 format (AV1 codec). The dataset also includes metadata such as task index, subtask labels, training flags, timestamps, frame indices, and episode indices. Data is stored in parquet files, with video files stored separately, totaling approximately 400MB (200MB for data and 200MB for videos). It is suitable for robotics learning, imitation learning, or reinforcement learning tasks. The dataset was created using the LeRobot project and is licensed under Apache 2.0.