HyeonseokE/SO101-Teleop-Open_drawer_40epi
收藏资源简介:
该数据集是一个用于机器人控制的开源数据集,专门针对打开抽屉的任务。它包含40个训练episodes,总计9364帧数据,使用SO_follower机器人类型。数据集提供了丰富的特征,包括动作(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)、观测状态(相同的6个关节位置),以及来自两个摄像头的图像观测:左腕摄像头和顶部摄像头,每个图像分辨率为480x640,3通道RGB,帧率为30fps。数据以Parquet文件格式存储,并包含时间戳、帧索引、episode索引等元信息。该数据集适用于机器人学习、强化学习或模仿学习的研究,特别是针对机械臂操作任务。
This dataset is an open-source dataset for robotics control, specifically focused on the task of opening a drawer. It contains 40 training episodes, totaling 9364 frames, and uses the SO_follower robot type. The dataset provides rich features, including actions (6 joint positions: shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper position), observation states (the same 6 joint positions), and image observations from two camera views: left wrist camera and top camera, each with a resolution of 480x640, 3-channel RGB, and a frame rate of 30fps. The data is stored in Parquet file format and includes metadata such as timestamps, frame indices, and episode indices. This dataset is suitable for research in robotics learning, reinforcement learning, or imitation learning, particularly for robotic arm manipulation tasks.




