danny1002/so101_move_box_left_20260528_134654
收藏资源简介:
该数据集由LeRobot创建,是一个机器人学数据集,专注于机械臂控制任务。数据集包含32个episodes,总计12606帧数据,采样率为25fps。数据特征包括:动作(6个浮点数,表示机械臂关节位置,如肩部平移、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(与动作相同的6个关节位置);观测图像(包括俯视图(overhead)和夹爪视图(gripper),均为480x640分辨率、3通道的RGB视频);以及时间戳、帧索引、episode索引等元数据。数据以parquet格式存储,总大小约300MB(数据文件100MB,视频文件200MB)。机器人类型为so_follower,适用于训练和评估机器人控制模型。
This dataset was created by LeRobot, and it is a robotics dataset focused on robotic arm control tasks. It contains 32 episodes, with a total of 12,606 frames and a sampling rate of 25 fps. The data features are as follows: actions, which are 6 floating-point numbers representing the robotic arm joint positions including shoulder translation, shoulder elevation, elbow flexion, wrist flexion, wrist rotation and gripper position; observation states, which are the same 6 joint positions as the actions; observation images, including overhead view and gripper view, both 3-channel RGB videos with a resolution of 480×640; and metadata such as timestamps, frame indices, episode indices and other related information. The data is stored in Parquet format, with a total size of approximately 300 MB (100 MB for data files and 200 MB for video files). The robotic platform type is so_follower, and this dataset is suitable for training and evaluating robotic control models.




