jungjaeuk/project_20260528_164308
收藏资源简介:
该数据集是一个用于机器人学研究的LeRobot数据集,通过LeRobot项目创建。数据集包含机器人动作和观测状态数据,具体特征包括6个关节位置(肩部平移、肩部升降、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置),这些数据以浮点32位格式存储。此外,数据集还包含时间戳、帧索引、episode索引、任务索引等元数据。数据集总共有1个episode、2397帧和1个任务,数据以parquet文件格式组织,分块存储(每块1000帧),总数据文件大小为100MB,视频文件大小为200MB。视频文件以mp4格式存储,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算法,支持机器人动作学习和状态观测任务。
This dataset is a LeRobot dataset for robotics research, created using the LeRobot project. It contains robot action and observation state data, with specific features including 6 joint positions (shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper positions), stored in float32 format. Additionally, the dataset includes metadata such as timestamps, frame indices, episode indices, and task indices. The dataset consists of 1 episode, 2397 frames, and 1 task, organized in parquet file format with chunked storage (1000 frames per chunk). The total data file size is 100MB, and video file size is 200MB, with video files stored in mp4 format. The robot type is so_follower. The dataset is primarily used for training and evaluating robot control algorithms, supporting tasks in robot action learning and state observation.




