Robot Autonomous Motion (RoAM) dataset
收藏资源简介:
RoAM数据集是由印度理工学院航空航天工程系创建的,主要用于自主机器人的室内人类运动预测。该数据集包含38000对立体图像,记录了定制的turtlebot3 Burger机器人在多种室内环境中的人类运动。数据集还包括同步的激光雷达扫描记录和机器人的所有控制动作。RoAM数据集的创建旨在解决自主移动机器人在部分可观测场景中的视觉预测问题,特别是在相机安装在移动平台上的情况下。该数据集的应用领域包括移动机器人和自主导航研究,旨在通过预测未来图像帧来优化机器人的运动规划。
The RoAM dataset was created by the Department of Aerospace Engineering, Indian Institute of Technology, and is primarily designed for indoor human motion prediction research for autonomous robots. This dataset contains 38,000 pairs of stereo images, which capture human motion data as the customized TurtleBot3 Burger robot navigates various indoor environments. It also includes synchronized LiDAR scan records and all control actions performed by the robot. The RoAM dataset was developed to solve visual prediction problems for autonomous mobile robots in partially observable scenarios, particularly when the camera is mounted on a mobile platform. Its application fields cover mobile robotics and autonomous navigation research, aiming to optimize robot motion planning by predicting future image frames.




