基于改进型蚁群算法的最优路径问题求解①.PDFVIP

计 算 机 系 统 应 用 2012 年 第 2 1 卷 第 10 期 基于改进型蚁群算法的最优路径问题求解① 张志协, 曹 阳 (华南师范大学 计算机学院, 广州 510631) 摘 要: 如何高效的向用户提供最优路径是蚁群算法大规模应用于导航系统的关键问题, 针对现有最优路径问 * 题研究中蚁群算法收敛速度慢及容易发生停滞的缺点, 利用 A 算法的启发式信息改进蚁群算法的路径选择策 略, 加快算法收敛速度. 同时引入遗传算法的双种群策略和蚁群系统信息素更新策略, 增加全局搜索能力, 避免 算法出现停滞现象. 仿真实验结果表明, 该改进算法具有较好的稳定性和全局优化性, 且收敛速度较快. * 关键词: 最优路径; 蚁群算法; A 算法; 双种群策略 Solving Optimal Path Problem Based on Improved Ant Colony Algorithm ZHANG Zhi-Xie, CAO Yang (School of Computer, South China Normal University, Guangzhou 510631, China) Abstract: Efficient optimal path is the key issue of the Ant Colony Algorithm used in road traffic navigation system. Aiming at the problem of slow convergence and stagnation phenomenon of Ant Colony Algorithm, this paper introduce path selection strategy which is based on the heuristic factor of A * algorithm to speed up the convergence. Meanwhile the Double-population strategy of the Genetic algorithm and the pheromone update rules of the Ant Colony System (ACS) are introduced in the algorithm, which avoid stagnation of the algorithm of the algorithm and speed up convergence. The results of the simulation experiment show that the improved algorithm has good stability, global optimization and fast convergence. Key words: optimal path; Ant Colony Algorithm; A * algorithm; Double-population strategy 最优路径问题是网络优化技术的核心, 自20 世纪 强的鲁棒性和发现较好解的能力, 但与此同时蚁群算 50 年代开始就陆续有各个领域的学者对网络最优路径 法也存在一些缺点, 如收敛速度慢、易出现停滞现象