grab_oranges_40ep_v1

该数据集是一个机器人学数据集，使用LeRobot工具创建，专注于机器人操作任务，例如抓取橙子。它包含21个总片段和10394个总帧，帧率为30 fps。数据集特征包括：机器人的动作（如肩部、肘部、手腕和夹爪的位置）、观测状态（如关节位置）、来自前置摄像头的图像观测（分辨率为480x640，RGB格式），以及时间戳、帧索引、片段索引等元数据。数据以Parquet格式存储，视频以MP4格式存储，总数据大小约为100 MB，视频文件大小约为200 MB。该数据集适用于机器人控制、强化学习或视觉导航任务的研究。This dataset is a robotics dataset created using the LeRobot tool, focusing on robot manipulation tasks such as grasping oranges. It contains a total of 21 episodes and 10394 frames with a frame rate of 30 fps. Features include robot actions (e.g., positions of shoulders, elbows, wrists, and grippers), observation states (e.g., joint positions), image observations from a front-facing camera (resolution 480x640, RGB format), and metadata such as timestamps, frame indices, and episode indices. The data is stored in Parquet format, while videos are in MP4 format, with total data size approximately 100 MB and video file size approximately 200 MB. This dataset is suitable for research in robot control, reinforcement learning, or visual navigation tasks.