whiteboard_inspect_0708

该数据集是使用LeRobot创建的机器人数据集，专门用于机器人学习任务。它包含一个双Flexiv Rizon4 RT机器人（bi_flexiv_rizon4_rt）的19个训练集，总帧数为67684，数据以30fps的帧率采集。数据特征包括动作（action）和观察（observation），其中动作和状态数据均为20维浮点数组，表示左右机械臂的TCP（工具中心点）位置、旋转和夹爪位置。观察数据包括头部、左右腕部和左右触觉传感器的视频图像（分辨率分别为480x640和400x700），以及时间戳、帧索引、集索引等元数据。数据集以Apache 2.0许可证发布，适用于机器人控制、模仿学习等研究。This robotic dataset is developed using LeRobot, specifically tailored for robotic learning tasks and research. It consists of 19 training splits based on a dual Flexiv Rizon4 RT robotic platform (identifier: bi_flexiv_rizon4_rt), with a total of 67,684 frames collected at a frame rate of 30 frames per second (fps). The dataset includes two primary data modalities: action and observation. Both action and state data are 20-dimensional floating-point arrays, which encode the position, rotation, and gripper position of the Tool Center Point (TCP) for both the left and right robotic arms. The observation data comprises video images captured by the head, left wrist, and right wrist cameras, as well as left and right tactile sensors, with resolutions of 480×640 and 400×700 respectively. Additionally, the dataset provides metadata including timestamps, frame indices, split indices and other relevant information. This dataset is released under the Apache 2.0 license, and is applicable to research areas such as robotic control and imitation learning.