stevenworkspace/eval_taken_5
收藏资源简介:
该数据集是一个机器人学习数据集,使用LeRobot工具创建,专门用于机器人控制和感知任务。数据集包含一个总剧集,共有3205帧数据,对应1个任务。数据以parquet文件格式存储,视频数据以mp4格式存储,帧率为30fps。数据集的特征包括:动作数据(16维浮点向量,表示左右机械臂关节位置和速度)、观测状态数据(与动作数据相同的16维向量)、来自三个摄像头的图像观测(高置摄像头、左手腕摄像头、右手腕摄像头,每个视频分辨率为480x640,3通道彩色视频),以及时间戳、帧索引、剧集索引、索引和任务索引等元数据。数据集适用于机器人强化学习、行为克隆和视觉-运动控制等研究。
This dataset is a robotics learning dataset created using the LeRobot tool, specifically designed for robot control and perception tasks. It contains one total episode with 3205 frames, corresponding to 1 task. The data is stored in parquet file format, with video data in mp4 format at 30fps. The dataset features include: action data (a 16-dimensional float vector representing left and right arm joint positions and velocities), observation state data (the same 16-dimensional vector as action data), image observations from three cameras (high camera, left wrist camera, right wrist camera, each with video resolution of 480x640, 3-channel color video), and metadata such as timestamp, frame index, episode index, index, and task index. The dataset is suitable for research in robot reinforcement learning, behavior cloning, and vision-motor control.




