ankile/square-d1-01a-baseline-uniform-r1-policy-rollouts

这是一个由LeRobot创建的机器人学习数据集，专门用于Panda机器人的操作任务。数据集包含200个训练集片段，总计58,983帧数据，帧率为20 fps。数据以Parquet文件格式存储，总数据文件大小为100 MB，视频文件大小为200 MB。数据集的特征包括机器人的观察状态（如末端执行器位置、姿态、夹爪位置）、动作（如位置和旋转增量、夹爪动作）、环境状态（如螺母和钉子的相对位置和姿态），以及任务相关的指标（如剩余步数、成功标志、奖励、完成标志等）。数据集主要用于训练机器人执行精确操作任务，例如装配或抓取。This is a robotics learning dataset created by LeRobot, specifically for Panda robot manipulation tasks. The dataset contains 200 training episodes, with a total of 58,983 frames at a frame rate of 20 fps. Data is stored in Parquet format, with a total data file size of 100 MB and video file size of 200 MB. Features include robot observation states (e.g., end-effector position, orientation, gripper positions), actions (e.g., position and rotation deltas, gripper actions), environment states (e.g., relative positions and orientations of nuts and pegs), and task-related metrics (e.g., steps to go, success flag, reward, done flag). The dataset is designed for training robots in precise manipulation tasks, such as assembly or grasping.