stevenworkspace/eval_take_mvactd_2
收藏资源简介:
该数据集是一个使用LeRobot创建的机器人数据集,专门用于移动AI机器人(mobileai_robot)的研究。数据集包含一个完整的情节(total_episodes: 1)和一个任务(total_tasks: 1),总帧数为1450。数据以parquet格式存储,视频以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,表示机器人左右关节的位置和速度;图像观测(observation.images)来自三个摄像头(cam_high、cam_left_wrist、cam_right_wrist),每个视频的分辨率为480x640,3通道彩色图像;此外,还包括时间戳、帧索引、情节索引、任务索引等元数据。该数据集适用于机器人控制、视觉导航和强化学习等任务。
This dataset is a robotics dataset created using LeRobot, specifically designed for mobile AI robot (mobileai_robot) research. It contains one complete episode (total_episodes: 1) and one task (total_tasks: 1), with a total of 1450 frames. The data is stored in parquet format, and videos are in mp4 format with a frame rate of 30fps. The dataset features include: action and observation.state, both as 16-dimensional float arrays representing the positions and velocities of the robots left and right joints; image observations (observation.images) from three cameras (cam_high, cam_left_wrist, cam_right_wrist), each with video resolution of 480x640 and 3-channel color images; additionally, it includes metadata such as timestamp, frame index, episode index, and task index. This dataset is suitable for tasks like robot control, visual navigation, and reinforcement learning.




