DORLR/teleop_isaaclab_ur5_udsci_success_16_demo
收藏资源简介:
该数据集是一个用于机器人学任务的数据集,专门针对UR5机器人执行变形堆叠长方体插入任务(环境标识:UR5-Deformable-StackedCuboid-Insertion-v0)。数据集使用LeRobot框架创建,包含2个完整episodes,总计2320帧,帧率为30fps。数据以parquet文件格式存储,总数据文件大小为100MB,视频文件大小为200MB。数据集特征包括:时间戳、帧索引、episode索引、任务索引;观察部分包括来自两个摄像头(cam_side2和cam_color)的图像,分辨率均为512x512x3,以及机器人的关节状态(7个关节角度)和笛卡尔状态(末端执行器位置和四元数姿态);动作部分包含7个关节的控制指令。数据集适用于机器人控制、强化学习或模仿学习等研究,标签信息表明其创建于2026年5月28日,运行在x86_64平台的机器人服务器上。
This dataset is tailored for robotics research, specifically targeting the deformable stacked cuboid insertion task executed by the UR5 robot (environment identifier: UR5-Deformable-StackedCuboid-Insertion-v0). It was constructed using the LeRobot framework, containing 2 complete episodes with a total of 2320 frames at a frame rate of 30 fps. The dataset is stored in Parquet file format, with a total data file size of 100 MB and a video file size of 200 MB. Dataset features include timestamps, frame index, episode index, and task index. The observation space comprises images captured by two cameras (cam_side2 and cam_color), both with a resolution of 512x512x3, as well as the robot’s joint states (7 joint angles) and Cartesian states (end-effector position and quaternion pose). The action space contains control commands for the 7 joints. This dataset is suitable for research in robot control, reinforcement learning, and imitation learning. The label information indicates that it was created on May 28, 2026, and runs on a robotic server with an x86_64 platform.




