mug_on_platform_20250821_subsampled_noWrist

这是一个使用LeRobot创建的机器人控制数据集，包含21个训练片段和5924帧数据。数据集提供了双机械臂系统的多模态数据，包括18维的关节状态观测和动作控制指令、10维的末端执行器位姿信息，以及来自Azure Kinect相机的720x1280分辨率的深度图像和彩色图像。数据以15fps的帧率采集，主要用于机器人学习和控制任务的研究。

This is a robotic control dataset developed using LeRobot, consisting of 21 training segments and a total of 5924 frames of data. The dataset provides multimodal data for a dual-manipulator system, including 18-dimensional joint state observations and action control commands, 10-dimensional end-effector pose information, as well as 720×1280 resolution depth and color images captured by an Azure Kinect camera. The data was collected at a frame rate of 15 fps, and is primarily intended for research on robotic learning and control tasks.