crislmfroes/openarm-pick-exhaust-pipe-sim-full-coverage

该数据集是一个用于机器人抓取任务的仿真数据集，由LeRobot创建，专注于OpenArm-Bimanual（双手臂机器人）在模拟环境中执行拾取排气管的任务。数据集包含465个完整的情节（episodes），总计100,551帧数据，以30帧每秒的速率采集。数据以视频和传感器数据形式存储，包括三个摄像头视角的观察图像：顶部摄像头、左腕摄像头和右腕摄像头，每个视频的分辨率为256x256像素，3通道（RGB）。此外，数据集还提供了机器人的状态观察（如左右手臂的位置、四元数姿态和夹爪状态）和动作数据（如左右手臂的位移、旋转和夹爪控制）。所有数据仅用于训练目的，没有单独的验证或测试分割。数据集采用Apache 2.0许可证。This dataset is a simulation dataset for robotic grasping tasks, created by LeRobot, focusing on the OpenArm-Bimanual (dual-arm robot) performing the task of picking up an exhaust pipe in a simulated environment. It contains 465 total episodes, with a total of 100,551 frames collected at 30 frames per second. The data is stored in video and sensor formats, including observation images from three camera perspectives: top camera, left wrist camera, and right wrist camera, each with a resolution of 256x256 pixels and 3 channels (RGB). Additionally, the dataset provides robot state observations (such as position, quaternion pose, and gripper state for both arms) and action data (such as displacement, rotation, and gripper control for both arms). All data is intended for training purposes only, with no separate validation or test splits. The dataset is licensed under Apache 2.0.