摘要
针对虚拟校园中道路环境存在U型槽及狭长路径的特点,提出一种基于改进遗传算法的路径规划方案。利用网格表示校园环境,并引入障碍物顶点信息,进行校园路径信息描述。在基本遗传算法进化过程中,加入平滑算子和避障算子,对适应值逐步收敛的种群进行人工选择优化,来提高成功率和适应值。实验结果表明,该方法能够在路径规划中准确的绕开U型槽障碍,并穿过狭长路径。
A path planning method based on a modified genetic algorithm for the U-shape and nar- row path in virtual campus is proposed. By using mesh to express the campus environment and introducing the vertex information of obstacles, the path information of campus is described. In progress of evolution of the basic genetic algorithm, smooth operator and obstacle avoidance op- erator are used to optimize the population by artificial selection, which its fitness gradually con- verges. Thus the shortcomings of low success rate and low fitness can be effectively improved. Experimental result shows that this method can achieve U-shape obstacles bypass and across nar- row path accurately.
出处
《西安邮电大学学报》
2013年第6期80-84,共5页
Journal of Xi’an University of Posts and Telecommunications
基金
西安市科技局技术转移促进工程基金资助项目(CXY1345-1)
西安市科技局技术转移促进工程基金资助项目(高校类)(CXY1339-6)
关键词
遗传算法
路径规划
漫游系统
虚拟校园
genetic algorithm, path planning, roaming system, virtual campus
