record-test

该数据集是一个专为机器人学习设计的数据集，由LeRobot工具创建。它包含5个训练集，总帧数为4156帧，帧率为30fps，适用于so101_follower型机器人，可能用于机械臂的跟随任务。数据集结构包括动作特征（由6个浮点数表示，对应机械臂关节位置，如肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置）、观察状态特征（与动作相同的6个关节位置）、前端摄像头图像（视频格式，分辨率为480x640，3通道RGB，无音频）以及元数据（如时间戳、帧索引和集索引）。这些特征支持机器人控制、模仿学习和强化学习任务，旨在通过整合视觉数据和关节位置信息来训练机器人执行精确动作。This dataset is designed for robot learning, created by the LeRobot tool. It contains 5 training sets with a total of 4156 frames at a frame rate of 30fps, suitable for the so101_follower robot type, potentially for robotic arm following tasks. The dataset structure includes action features (6 floating-point numbers representing joint positions of the robotic arm, such as shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position), observation state features (the same 6 joint positions as actions), front camera images (video format, resolution 480x640, 3-channel RGB, no audio), and metadata (such as timestamps, frame indices, and episode indices). These features are applicable to robot control, imitation learning, or reinforcement learning tasks, aiming to train robots to perform actions through visual and joint position data.