Jiafei1224/19042026-pipette-tip-combined

该数据集是一个机器人学习数据集，专为机器人控制任务设计，使用LeRobot工具创建。它包含200个episodes和141,748帧数据，帧率为30 fps。数据集结构包括动作（action）和观察（observation），其中观察部分涵盖状态（如关节位置）和图像（来自右、左和顶部摄像头的视频，分辨率为360x640，3通道RGB）。机器人类型为bi_yam_follower，可能表示一个双臂机器人系统。数据以parquet文件格式存储，并附带视频文件。数据集适用于机器人模仿学习、强化学习等研究，但具体任务细节（如任务名称）未在README中明确说明。This is a robotics learning dataset developed for robotic control tasks, built using the LeRobot toolkit. It comprises 200 episodes and 141,748 frames of data, with a frame rate of 30 fps. The dataset structure includes action and observation modalities: the observation modality covers state information (e.g., joint positions) and image data, specifically videos captured by the right, left, and top cameras, which are 3-channel RGB images with a resolution of 360×640. The robot platform is bi_yam_follower, which presumably denotes a dual-arm robotic system. The dataset is stored in Parquet file format, with accompanying video files. This dataset is applicable to research in robotic imitation learning, reinforcement learning and other related domains, yet the specific task details (such as the task name) are not explicitly stated in the README.