HyeonseokE/SO101-cap_open_pot_10fps_80epi
收藏资源简介:
该数据集由LeRobot创建,是一个用于机器人学任务的数据集,特别针对SO101机器人(follower类型)。数据集包含80个完整的情节(episodes),总计27008帧数据,帧率为10fps。数据以Parquet格式存储,总数据文件大小为100MB,视频文件大小为200MB。数据集特征丰富,包括机器人状态观测(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)、动作(相同的6个关节位置)、来自顶部和左腕摄像头的图像观测(分辨率480x640,RGB三通道,H.264编码视频)、末端执行器位置(XYZ坐标和欧拉角)、夹爪二进制状态、技能信息(自然语言描述、验证问题、类型、进度、目标位置)、子任务信息(自然语言描述、对象名称、目标位置)以及时间戳、帧索引、情节索引等元数据。数据集仅包含训练集,覆盖一个任务,适用于机器人控制、模仿学习或强化学习研究。
This dataset was created using LeRobot and is designed for robotics tasks, specifically for the SO101 robot (follower type). It contains 80 total episodes with 27,008 frames at a frame rate of 10 fps. The data is stored in Parquet format, with a total data file size of 100 MB and video file size of 200 MB. The dataset includes rich features such as robot state observations (6 joint positions: shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper positions), actions (same 6 joint positions), image observations from top and left wrist cameras (resolution 480x640, RGB three-channel, H.264 encoded video), end-effector positions (XYZ coordinates and Euler angles), gripper binary state, skill information (natural language description, verification question, type, progress, goal positions), subtask information (natural language description, object name, target position), and metadata like timestamp, frame index, episode index. The dataset only includes a training split, covers one task, and is suitable for robotics control, imitation learning, or reinforcement learning research.




