merged-record-test_3

该数据集由LeRobot创建，专注于机器人学领域。数据集包含192个训练片段，总计70,002帧数据，涉及1个任务。数据采用多模态格式，包括机器人的动作数据（6个关节位置：肩部平移、肩部抬升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置）、状态观测（与动作相同的关节位置）、以及三个视角的视频观测（前置摄像头、腕部摄像头和上方摄像头），视频分辨率为480x640，帧率为30fps，编码格式为h264。数据集还包含时间戳、帧索引、片段索引等时序信息，适用于机器人控制与模仿学习任务。This dataset was created by LeRobot and focuses on the field of robotics. It contains 192 training episodes, with a total of 70,002 frames, covering one single task. The data is in multimodal format, including robot action data (6 joint positions: shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position), state observations (the same joint positions as the action data), and video observations from three perspectives: front-facing camera, wrist-mounted camera, and overhead camera. The videos have a resolution of 480x640, a frame rate of 30 fps, and are encoded in h264 format. The dataset also includes temporal information such as timestamps, frame indices, and episode indices, which is suitable for robot control and imitation learning tasks.