stevenworkspace/eval_take_mvactd_3
收藏资源简介:
该数据集是使用LeRobot创建的机器人数据集,专注于移动AI机器人(mobileai_robot)任务。数据集包含一个完整的episode,总计1082帧,对应一个任务。数据以parquet文件格式存储,视频文件以mp4格式存储,帧率为30fps。数据集的特征包括:动作(action)和观测状态(observation.state),均为16维浮点数组,表示左右机械臂的关节位置(如left_joint_0.pos到right_joint_5.pos)以及线速度和角速度(x.vel, theta.vel);图像观测来自三个摄像头:cam_high、cam_left_wrist和cam_right_wrist,每个视频的分辨率为480x640,3通道彩色,使用av1编解码器;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。数据集适用于机器人学习和控制研究,特别是多模态感知和动作生成任务。
This dataset is a robotics dataset created using LeRobot, focusing on mobile AI robot (mobileai_robot) tasks. It contains one total episode with 1082 frames, corresponding to a single task. The data is stored in parquet file format, with video files in mp4 format at 30fps. The dataset features include: action and observation.state, both as 16-dimensional float32 arrays representing joint positions of left and right robotic arms (e.g., left_joint_0.pos to right_joint_5.pos) and linear and angular velocities (x.vel, theta.vel); image observations from three cameras: cam_high, cam_left_wrist, and cam_right_wrist, each video with a resolution of 480x640, 3 color channels, using the av1 codec; additionally, metadata such as timestamp, frame index, episode index, index, and task index are included. The dataset is suitable for robotics learning and control research, particularly for multimodal perception and action generation tasks.




