摘要
未来智能导航机器人高端产业将是一块热门且发展迅速的领域,基于视觉的同时定位与建图(V-SLAM)便是智能导航机器人的热门研究方向之一。本文设计了一款基于Arduino平台的循迹小车,其搭载双目相机并运用ORB-SLAM算法建立即时地图。通过数学建模及假设分析,判断其在实际使用过程中完成具体功能的可行性。通过小车行驶试验仿真分析,选择最适合的算法程序与硬件设施,探究并解决这种理论构想的导航小车在实际运行中所存在的问题,并对其特点进行评析。经过改良的导航小车不仅能精确行驶循迹路线,还能在非循迹情况下实现自动导航的功能,并实现高精度的场景重建与避障功能。
In the future,the high-end industry of intelligent navigation robots will be a hot and rapidly developing field.Vision based simultaneous localization and mapping(V-SLAM)is one of the hot research directions of intelligent navigation robots.A tracking car based on Arduino platform was designed,which was equipped with a binocular camera and used ORB-SLAM algorithm to create a real-time map.Through mathematical modeling and hypothesis analysis,the feasibility of completing specific functions in the actual use process was judged.Through the simulation analysis of the car running experiment,the most suitable algorithm program and hardware facilities were selected to explore and solve the problems in the actual operation of the navigation car conceived in this theory,and its characteristics were evaluated.The improved navigation car can not only accurately follow the route,but also realize the function of automatic navigation under non tracking conditions,and realize high-precision scene reconstruction and obstacle avoidance.
作者
贾凡
蒲沛
童德林
农胜旺
JIA Fan;PU Pei;TONG Delin;NONG Shengwang(School of Automotive and Traffic Engineering,Jiangsu University,Zhenjiang Jiangsu 212013,China)
出处
《汽车零部件》
2022年第12期21-24,共4页
Automobile Parts
基金
江苏大学2021年大学生实践创新训练计划项目(202110299171Y)。
