record-over_20260525_183452

该数据集是一个用于机器人学习的数据集，创建于LeRobot平台。数据集包含50个episodes，总计14943帧，数据以parquet格式存储，视频文件为MP4格式，帧率为30fps，图像分辨率为480x640。特征包括机器人动作（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（与动作相同的关节位置）以及前端摄像头图像。此外，还包含时间戳、帧索引、episode索引、任务索引等元数据。机器人类型为“so_follower”，数据集大小为100MB（数据文件）和200MB（视频文件），并分为训练集（涵盖所有episodes）。This dataset is designed for robot learning and was created on the LeRobot platform. It contains 50 episodes, totaling 14943 frames, with data stored in parquet format and video files in MP4 format at a frame rate of 30fps and an image resolution of 480x640. Features include robot actions (such as shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation, and gripper position), observation states (joint positions identical to the actions), and front camera images. Additionally, it includes metadata such as timestamps, frame indices, episode indices, and task indices. The robot type is so_follower, with a dataset size of 100MB (data files) and 200MB (video files), and it is divided into a training set (covering all episodes).