KS325/open-upper-drawer-r2-occ-train-expand

该数据集是一个机器人学习数据集，专门用于训练机器人执行打开上层抽屉的任务。数据集由LeRobot平台创建，使用so_follower类型机器人收集。包含28个完整的episodes，总计20407帧数据，帧率为30fps。数据以parquet文件格式存储，总数据文件大小为100MB，视频文件大小为200MB。数据集特征包括：动作数据（6个浮点数值，对应机器人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）、状态观测（与动作相同的关节位置）、两个摄像头观测（camera1和camera2，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。所有数据仅用于训练分割，适用于机器人控制、模仿学习或强化学习研究。

This dataset is a robotics learning dataset specifically designed for training robots to perform the open upper drawer task. It was created using the LeRobot platform and collected with a so_follower type robot. The dataset contains 28 complete episodes, totaling 20,407 frames at a frame rate of 30fps. Data is stored in parquet format, with a total data file size of 100MB and video file size of 200MB. Features include: action data (6 float32 values corresponding to robot joint positions such as shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper position), state observation (same joint positions as actions), two camera observations (camera1 and camera2, both as 480x640 resolution, 3-channel RGB videos), and metadata such as timestamp, frame index, and episode index. All data is allocated for the training split, suitable for robotics control, imitation learning, or reinforcement learning research.