cell1_20260524_mohamed-salah_grocery_bagging_cycle_20260524_100832

该数据集是一个用于机器人技术研究的数据集，专门针对双机械臂机器人（starpilot_yam_gripper）的操作任务设计。数据集包含5个完整的episodes，总计1844帧数据，覆盖1个具体任务（例如物品抓取或操作）。数据以30 FPS的帧率采集，包括动作数据（14维浮点数组，代表两个机械臂的位置、旋转和夹爪宽度）、观测数据（32维浮点数组，包括编码器角度、IMU传感器数据和姿态信息）以及多摄像头图像观测（来自左腕、右腕和基座摄像头的RGB和深度图像，分辨率分别为480x640和768x1024）。所有数据以Parquet格式存储，并附带视频文件。该数据集适用于机器人学习、模仿学习、强化学习等研究，旨在帮助开发和控制双机械臂系统。数据集使用Apache 2.0许可证发布。This dataset is intended for robotics research, specifically designed for manipulation tasks of the dual-arm robot (starpilot_yam_gripper). It contains 5 full episodes, totaling 1844 frames of data covering one specific task, such as object grasping or manipulation. All data is collected at 30 FPS, including action data (a 14-dimensional floating-point array representing the positions, rotations, and gripper widths of the two robotic arms), observation data (a 32-dimensional floating-point array encompassing encoder angles, IMU sensor data, and pose information), and multi-camera image observations (RGB and depth images from the left wrist, right wrist, and base-mounted cameras, with resolutions of 480x640 and 768x1024 respectively). All data is stored in Parquet format, with accompanying video files. This dataset is suitable for research in robot learning, imitation learning, reinforcement learning and other related fields, aiming to facilitate the development and control of dual-arm robotic systems. The dataset is released under the Apache 2.0 license.