cell1_20260520_laundry_folding_20260520_125644

该数据集是一个使用LeRobot工具创建的机器人操作数据集，专为机器人学习任务设计。它包含2个完整的情节（episodes），共计1716帧数据，涉及一个推测为衣物折叠的具体任务（根据文件名推断）。数据采集自starpilot_yam_gripper型号的双机械臂机器人。数据特征包括：14维的动作空间（控制两个机械臂的末端执行器位置、旋转和夹爪宽度）；32维的观测状态（包括两个机械臂的编码器角度、IMU加速度和角速度、末端执行器位姿（位置和四元数）、夹爪开合状态和距离）；以及来自多个摄像头的视觉观测：左腕和右腕的RGB图像和深度图像（分辨率480x640），基座的两个RGB摄像头图像（一个分辨率480x640，另一个分辨率768x1024）。所有视频数据以30fps录制，使用h264编解码器。数据集以Apache 2.0许可证发布，主要用于机器人模仿学习或强化学习的研究和开发。This robotic manipulation dataset was constructed using the LeRobot toolkit, and is tailored for robotic learning tasks. It consists of 2 full episodes, totaling 1716 frames of data, corresponding to a specific task inferred to be cloth folding based on the filenames. Data was collected from a dual-arm robotic platform with the starpilot_yam_gripper model. The dataset's features include: a 14-dimensional action space that controls the positions, rotations, and gripper widths of the end-effectors of both robotic arms; a 32-dimensional observation state that encompasses encoder angles of both robotic arms, IMU acceleration and angular velocity, end-effector poses (position and quaternion), as well as gripper opening status and distance; plus visual observations from multiple cameras: RGB and depth images from the left and right wrist cameras (resolution 480×640), and two RGB camera images from the robot base, with resolutions of 480×640 and 768×1024 respectively. All video data is recorded at 30 frames per second (fps) using the H.264 codec. This dataset is released under the Apache 2.0 license, and is primarily intended for research and development in robotic imitation learning or reinforcement learning.