kunhsiang/eval_exp9_sc1var_grasp_the_big_box_20260528-184426
收藏资源简介:
该数据集是使用LeRobot创建的机器人数据集,专门用于机器人任务,涉及一个名为so_follower的机器人类型。数据集包含1个episode,总计2566帧,以30帧每秒的速率采集。数据特征包括动作(6个关节位置:肩部旋转、肩部提升、肘部弯曲、手腕弯曲、手腕滚动和夹爪位置)、观测状态(同样为6个关节位置),以及三个视角的图像观测:前视、顶部和夹爪视角,每个图像分辨率为480x640像素,3个通道,采用AV1编解码器。此外,数据集还包括时间戳、帧索引、episode索引、索引和任务索引等元数据。数据以parquet文件格式存储,视频以mp4格式存储,总数据文件大小为100MB,视频文件大小为200MB。
This dataset was created using LeRobot and is designed for robotics tasks, involving a robot type named so_follower. It contains 1 episode with a total of 2566 frames, captured at 30 frames per second. The features include actions (6 joint positions: shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper positions), observation states (also 6 joint positions), and image observations from three perspectives: front, top, and gripper, each with a resolution of 480x640 pixels, 3 channels, using the AV1 codec. Additionally, the dataset includes metadata such as timestamp, frame index, episode index, index, and task index. Data is stored in parquet files, with videos in mp4 format, totaling 100MB for data files and 200MB for video files.




