Jiafei1224/20042026-cup-storing-combined

该数据集是使用LeRobot创建的机器人学数据集，专门用于机器人控制与学习任务。数据集包含200个训练片段（episodes），总计167,327帧，采样频率为30fps。机器人类型为bi_yam_follower（可能表示双Yam机械臂跟随系统），任务数量为1。数据集结构包括动作数据（14维浮点数组，表示左右机械臂各6个关节及夹爪的位置）、状态观测数据（与动作相同的14维关节位置）、图像观测数据（来自右、左、顶部三个摄像头的视频流，分辨率为360x640，编码为AV1，无音频），以及时间戳、帧索引、片段索引、索引和任务索引等元数据。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。该数据集适用于机器人强化学习、模仿学习或行为克隆等研究，尤其适用于多视角视觉与关节控制结合的算法开发。This dataset is a robotics dataset created using LeRobot, designed for robot control and learning tasks. It contains 200 training episodes with a total of 167,327 frames at 30fps. The robot type is bi_yam_follower (likely indicating a dual Yam robotic arm following system), and there is 1 task. The dataset structure includes action data (14-dimensional float arrays representing positions of 6 joints and gripper for both left and right arms), state observation data (14-dimensional joint positions identical to actions), image observation data (video streams from right, left, and top cameras with 360x640 resolution, AV1 codec, no audio), and metadata such as timestamp, frame index, episode index, index, and task index. Data is stored in parquet format, with total data file size of 100MB and video file size of 200MB. This dataset is suitable for research in robot reinforcement learning, imitation learning, or behavior cloning, particularly for algorithm development combining multi-view vision and joint control.