KS325/open-lower-drawer-r2-occ-val-spare

该数据集是一个机器人学数据集，专为机器人控制和模仿学习任务设计，使用LeRobot工具创建。数据集包含4个完整的情节（episodes），总计2975帧数据，涉及1个具体任务。机器人类型为so_follower，数据以parquet文件格式组织，总数据文件大小为100MB，视频文件大小为200MB，帧率为30fps。数据集的特征包括：动作（action）和观测状态（observation.state），均为6维浮点数组，对应机器人的肩部平移、肩部抬升、肘部弯曲、腕部弯曲、腕部旋转和夹爪位置；观测图像来自两个摄像头（camera1和camera2），每个摄像头提供480x640分辨率、3通道（RGB）的视频数据，编码为av1格式；此外，还包括时间戳、帧索引、情节索引、索引和任务索引等元数据。数据集主要用于训练和评估机器人模仿学习模型，支持多模态输入（如关节状态和视觉信息）。

This dataset is a robotics dataset designed for robot control and imitation learning tasks, created using the LeRobot tool. It contains 4 complete episodes, totaling 2975 frames, and involves 1 specific task. The robot type is so_follower, and the data is organized in parquet file format, with a total data file size of 100MB and video file size of 200MB, at a frame rate of 30fps. The dataset features include: action and observation.state, both as 6-dimensional float arrays corresponding to the robots shoulder pan, shoulder lift, elbow flex, wrist flex, wrist roll, and gripper positions; observation images from two cameras (camera1 and camera2), each providing video data at 480x640 resolution with 3 channels (RGB), encoded in av1 format; additionally, it includes metadata such as timestamp, frame index, episode index, index, and task index. The dataset is primarily used for training and evaluating robot imitation learning models, supporting multimodal inputs (e.g., joint states and visual information).