edmos7/so101_pickplace_front_wrist_camera_yellow-trajectory

该数据集是一个机器人操作数据集，使用LeRobot创建，专门用于机器人学习任务。数据集包含60个总集数，18000个总帧数，帧率为30fps。数据以parquet文件格式存储，总数据文件大小为100MB，视频文件大小为200MB。数据集采用Apache 2.0许可证。特征包括动作（包含6个关节位置：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（与动作相同的关节位置）、时间戳、帧索引、集索引、索引和任务索引。观测图像包括主摄像头（MAIN_CAM）和副摄像头（SEC_CAM）的视频数据，分辨率均为480x640x3，使用AV1编解码器，以及轨迹图像（trajectory），分辨率为480x640x3。数据集适用于训练和评估机器人控制模型，特别是基于视觉的抓取和放置任务。This dataset is a robotics manipulation dataset created using LeRobot, specifically designed for robot learning tasks. It contains a total of 60 episodes, 18000 frames, with a frame rate of 30fps. The data is stored in parquet format, with a total data file size of 100MB and video file size of 200MB. The dataset is licensed under Apache 2.0. Features include action (with 6 joint positions: shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper position), observation state (same joint positions as action), timestamp, frame index, episode index, index, and task index. Observation images include video data from the main camera (MAIN_CAM) and secondary camera (SEC_CAM), both with a resolution of 480x640x3 using AV1 codec, as well as trajectory images (trajectory) with a resolution of 480x640x3. The dataset is suitable for training and evaluating robot control models, particularly for vision-based pick-and-place tasks.