burger-to-plate-320x240-filteredX

该数据集使用LeRobot创建，是一个机器人演示数据集，专门用于机器人学任务。数据集包含blueberry_ros类型机器人的55个episodes，总计86277帧，涉及1个任务。数据以parquet文件格式存储，视频文件总大小为200MB，帧率为15fps。特征包括动作（26维浮点数，涵盖左右手臂的线性速度、角速度、手指位置和基座控制）、观测状态（55维浮点数，包括关节位置、力矩以及gaze坐标和有效性）、多个图像观测（左、右、用户和用户gaze视角的240x320分辨率RGB视频，编码为av1，无音频），以及时间戳、帧索引、episode索引等元数据。数据集适用于机器人控制、模仿学习或视觉导航等研究。

This dataset is developed based on LeRobot, serving as a robotic demonstration dataset specialized for robotics-related tasks. It contains 55 episodes of the blueberry_ros type robot, totaling 86277 frames and covering 1 single task. The dataset is stored in Parquet file format, with the total size of video files reaching 200 MB and a frame rate of 15 fps. The included features are as follows: actions (26-dimensional floating-point vectors, covering linear velocities, angular velocities, finger positions, and base control for both left and right arms), observation states (55-dimensional floating-point vectors, including joint positions, joint torques, gaze coordinates and their validity), multiple image observations (240×320 resolution RGB videos captured from left, right, user, and user gaze perspectives, encoded in AV1 format without audio), as well as metadata such as timestamps, frame indices, and episode indices. This dataset is suitable for research in robot control, imitation learning, visual navigation, and other related robotics fields.