CV-Rob/rollout_so101_lego_pick_place_20260526_185349
收藏资源简介:
该数据集是使用LeRobot创建的机器人操作数据集,专注于乐高拾放任务。数据集包含2个总episodes和53242个总帧,帧率为30 fps。特征包括:动作(6个关节位置,如肩部平移、肩部提升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置)、观测状态(与动作相同的6个关节位置)、观测图像(来自顶部摄像头和夹爪摄像头的视频,分辨率480x640,3通道,编码为av1)、干预标志(布尔值)、时间戳、帧索引、episode索引、索引和任务索引。数据集结构为Parquet文件,总数据大小约100 MB,视频文件大小约200 MB,机器人类型为so_follower。适用于机器人学习任务,如模仿学习或强化学习。
This dataset is a robotics manipulation dataset created using LeRobot, focusing on a Lego pick-and-place task. It contains a total of 2 episodes and 53,242 frames with a frame rate of 30 fps. Features include: action (6 joint positions such as shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper position), observation state (same 6 joint positions as action), observation images (videos from a top camera and a gripper camera, resolution 480x640, 3 channels, encoded as av1), intervention flag (boolean), timestamp, frame index, episode index, index, and task index. The dataset is structured in Parquet files, with a total data size of approximately 100 MB and video file size of about 200 MB. The robot type is so_follower. It is suitable for robotics learning tasks such as imitation learning or reinforcement learning.




