usvz_cube_20260618_183728

该数据集是一个用于机器人技术领域的数据集，由LeRobot创建。它包含15个总剧集和15719个总帧，数据以parquet文件格式存储，视频文件为mp4格式。数据集的特征包括动作（action）、观察状态（observation.state）、前视图像（observation.images.front）、时间戳、帧索引、剧集索引、索引和任务索引。动作和观察状态均涉及6个关节位置，具体为肩部平移位置、肩部升降位置、肘部弯曲位置、腕部弯曲位置、腕部旋转位置和夹爪位置。前视图像为视频数据，分辨率为480x640，3通道（彩色），帧率为30fps，使用av1编解码器。机器人类型为so_follower。数据集主要用于训练和评估机器人控制模型，支持从状态和图像观测中学习动作策略。

This is a robotics dataset created by LeRobot. It contains 15 total episodes and 15719 total frames. The data is stored in Parquet file format, with video files in MP4 format. The dataset features include action, observation.state, front-facing images (observation.images.front), timestamps, frame indices, episode indices, index, and task index. Both action and observation.state involve 6 joint positions, specifically shoulder translation position, shoulder lift position, elbow flexion position, wrist flexion position, wrist rotation position, and gripper position. The front-facing images are video data with a resolution of 480×640, 3-channel (color), a frame rate of 30 fps, and encoded using the AV1 codec. The robot type is so_follower. This dataset is primarily used for training and evaluating robotic control models, supporting the learning of action policies from both state and image observations.