基于视觉的移动机器人实时避障和导航

阐述了移动机器人通过视觉传感器在不确定的环境中实现自主避障和导航的一种方法. 首先讨论了路径规划的分层结构,然后通过简单的图像处理方法获取运动环境中的避碰点,最后给出机器人实现局部路径规划的算法. 该方法有效地利用了全局环境地图和视觉信息,减少了计算量,提高实时性,通过仿真研究说明了该算法的有效性和可行性.
关键词:自主移动机器人;视觉系统;避　碰;路径规划
Abstract :The method of local path planning and navigation of mobile robot based on single visual information is in2 troduced in the paper. Firstly , the general concept of the hierarchical planning system is described. Then a method to observe the environment by single vision and a method to make an avoiding point are explained. Finally , a local path planning algorithm based on visual information is proposed . This method makes effeccive use of the vision in2
formation , reduces the scale of calculation and improves the celerity. The validity and feasibility of the method was confirmed by the simulation results and experiments.
Key words :mobile robot ;visual system; obstacle avoiding ; path planning