cell1_20260524_mohamed-salah_grocery_bagging_cycle_20260524_100641

该数据集是一个机器人学习数据集，由LeRobot工具创建。它包含5个episodes和1656帧数据，涉及一个任务（可能为杂货装袋）。数据集针对双机械臂系统（starpilot_yam_gripper机器人类型），包括动作数据（如14维的机械臂位置、旋转和夹爪宽度）、状态观测数据（如32维的编码器角度、IMU测量和姿态信息）以及多个摄像头图像观测（如左腕、右腕和基座摄像头的RGB图像，部分带深度信息）。数据以parquet文件和视频格式存储，帧率为30fps，适用于机器人控制、模仿学习或强化学习任务。This dataset is a robotic learning dataset developed by the LeRobot toolkit. It contains 5 episodes and 1656 frames of data, corresponding to a task (likely grocery bagging). The dataset is targeted at a dual robotic arm system (starpilot_yam_gripper robot type), and includes action data (e.g., 14-dimensional robotic arm position, rotation and gripper width), state observation data (e.g., 32-dimensional encoder angles, IMU measurements and pose information), as well as multiple camera image observations (e.g., RGB images from left wrist, right wrist and base-mounted cameras, some of which include depth information). The data is stored in Parquet files and video formats, with a frame rate of 30 fps. It is applicable to robotic control, imitation learning or reinforcement learning tasks.