Octa7000/eval_act_pick_cube_hand
收藏资源简介:
该数据集是一个机器人学领域的数据集,使用LeRobot工具创建。它包含10个训练集,总帧数为12654,涉及1个任务。数据以Parquet格式存储,总大小为100MB,视频文件大小为200MB,帧率为30fps。数据集的特征包括动作(action)和观察(observation),其中动作和状态观察均包含6个浮点型关节位置(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)。观察部分还包括来自前摄像头和腕部摄像头的视频图像,分辨率为480x640,3通道,采用AV1编解码器。此外,数据集还包含时间戳、帧索引、集索引、索引和任务索引等元数据。机器人类型为“so_follower”。
This dataset is a robotics dataset created using the LeRobot tool. It contains 10 training episodes with a total of 12654 frames and involves 1 task. The data is stored in Parquet format with a total size of 100MB, video files size of 200MB, and a frame rate of 30fps. The dataset features include action and observation, where both action and state observation consist of 6 floating-point joint positions (e.g., shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper position). The observation also includes video images from front and wrist cameras with a resolution of 480x640, 3 channels, using the AV1 codec. Additionally, the dataset contains metadata such as timestamp, frame index, episode index, index, and task index. The robot type is so_follower.



