LzDIY/so101_clip_box_onto_green_plate

该数据集是使用LeRobot创建的机器人数据集，专为机器人技术任务设计。数据集包含50个episodes，总计41998帧，涉及一个单一任务。数据以parquet文件格式存储，并包含相应的视频文件（总大小约200MB）。数据集的特征包括动作（action，包含6个关节位置：肩部平移、肩部提升、肘部弯曲、手腕弯曲、手腕旋转和夹爪位置）、观测状态（observation.state，同样包含6个关节位置）、头部图像（observation.images.head，分辨率为480x640的RGB视频，帧率30fps）和手腕图像（observation.images.wrist，分辨率相同）。此外，还包括时间戳、帧索引、episode索引、索引和任务索引等元数据。机器人类型为so_follower，数据分块大小为1000帧，训练集覆盖所有episodes。该数据集适用于机器人控制、视觉感知和强化学习等研究。

This dataset was created using LeRobot and is designed for robotics tasks. It contains 50 episodes, totaling 41,998 frames, and involves a single task. The data is stored in parquet format and includes corresponding video files (total size approximately 200MB). The dataset features include actions (with 6 joint positions: shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper position), observation states (similar joint positions), head images (observation.images.head, RGB video with 480x640 resolution at 30fps), and wrist images (observation.images.wrist, same resolution). Additionally, it includes metadata such as timestamp, frame index, episode index, index, and task index. The robot type is so_follower, with a chunk size of 1000 frames, and the training split covers all episodes. This dataset is suitable for research in robot control, visual perception, and reinforcement learning.