ecappiell/multiple_box_experiment_20260527_184122
收藏资源简介:
该数据集是一个机器人数据集,使用LeRobot创建。它包含机器人动作和观测数据,具体特征包括:动作数据(6个浮点值,对应肩部平移、肩部升降、肘部弯曲、腕部弯曲、腕部滚动和夹爪的位置);观测状态数据(117个浮点值,涵盖关节位置、原始电流、负载、硬件速度、速度、原始扭矩、签名电流、签名扭矩、估计扭矩、模型扭矩等);时间戳、帧索引、episode索引、索引和任务索引。此外,数据集提供两个视频观测:顶部图像和角落图像,均为480x640分辨率、3通道、30fps、使用av1编解码器。数据集总共有1个episode、630帧、1个任务,数据以parquet格式存储,机器人类型为so_follower,适用于机器人学习和控制任务。
This dataset is a robotics dataset created using LeRobot. It includes robot action and observation data, with specific features: action data (6 float values for positions of shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper); observation state data (117 float values covering joint positions, raw current, load, hardware velocity, velocity, raw torque, signed current, signed torque, estimated torque, model torque, etc.); and timestamps, frame index, episode index, index, and task index. Additionally, the dataset provides two video observations: top images and corner images, both with 480x640 resolution, 3 channels, 30fps, using av1 codec. The dataset totals 1 episode, 630 frames, and 1 task, stored in parquet format, with robot type so_follower, suitable for robotics learning and control tasks.




