gsethia08/part4-screw_unscrew_bottle_cap

该数据集是使用LeRobot工具创建的，专注于机器人技术领域，特别是手部机器人操作。数据集包含2868个episodes和370834个frames，覆盖20个不同任务。数据以30fps的帧率采集，包括视频观察（分辨率为1080x1920的RGB视频）、机器人状态观察（48维浮点数组，表示左右手腕和手指的位置与旋转信息）、相机内外参数、动作数据（48维浮点数组）、语言指令、时间戳、帧索引、episode索引、任务索引等特征。数据集主要用于训练和评估机器人学习模型，支持手部机器人的操作任务，如拧瓶盖等。数据以Parquet格式存储，总数据文件大小为100MB，视频文件大小为200MB。This dataset was developed using the LeRobot toolkit, focusing on the robotics domain, particularly dexterous robotic manipulation tasks. It contains 2,868 episodes and 370,834 frames, covering 20 distinct tasks. All data is collected at a frame rate of 30 frames per second (fps), and includes multiple modalities: video observations (RGB videos with a resolution of 1080×1920), robotic state observations (48-dimensional floating-point arrays encoding the position and rotation information of the left and right wrists and fingers), camera intrinsic and extrinsic parameters, action data (48-dimensional floating-point arrays), language instructions, timestamps, frame indices, episode indices, task indices, and other associated metadata. This dataset is primarily used for training and evaluating robotic learning models, supporting dexterous robotic manipulation tasks such as unscrewing bottle caps. The dataset is stored in Parquet format, with a total data file size of 100 MB and a total video file size of 200 MB.