摘要
为了改善传统蚁群算法在路径规划中缺乏足够鲁棒性的问题,釆用改进自适应蚁群算法,根据解的分布情况自适应地进行信息素的更新,在多种步长选择机制下选择最优步长,提高全局搜索能力。在MATLAB中对本文算法、传统蚁群算法以及自适应蚁群算法分别进行了仿真实验对比。在相同的环境模型下,该算法的迭代次数为2次,比自适应蚁群算法提升了93%,最小路径长度为27.67,比自适应蚁群算法提升了3.4%;在给定的复杂环境模型下进行路径规划时,该算法的迭代次数为2次,最小路径长度为28.88,传统蚁群算法对应的迭代次数和路径长度分别为166和29.8。仿真结果表明,改进后的蚁群算法较传统蚁群算法相比,能够快速找到最短路径,并具有更好的稳定性和收敛性。
In order to improve the robustness of traditional ant colony algorithm in path planning,the improve adaptive ant colony algorithm is used to update pheromones adaptively according to the distribution of solutions,and the optimal step size is selected under various step size selection mechanisms so as to improve the global search ability.In MATLAB,the algorithm,the traditional ant colony algorithm and adaptive ant colony algorithm are simulated and compared.Under the same environment model,the number of iterations of this algorithm is 2 times,which is 93%higher than adaptive ant colony algorithm,and the minimum path length is 27.67,which is 3.4%higher than adaptive ant colony algorithm.In the given complex environment model,the number of iterations of this algorithm is 2,the minimum path length is 2&88,and the number of iterations and path length corresponding to the traditional ant colony algorithm are 166 and 29.8,respectively.The simulation results show that the improved ant colony algorithm can find the shortest path faster than the traditional ant colony algorithm,and has better stability and convergence.
作者
徐玉琼
娄柯
李婷婷
高文根
Xu Yuqiong;Lou Ke;Li Tingting;Gao Wengen(College of Electrical Engineering,Anhui Polytechnic University,Wuhu 241000,China;Key Laboratory of Detection Technology and Energy Saving Devices of Anhui Province,Wuhu 241000,China)
出处
《电子测量与仪器学报》
CSCD
北大核心
2019年第10期89-95,共7页
Journal of Electronic Measurement and Instrumentation
基金
安徽省高校自然科学研究重点项目(KJ2019A0151,KJ2019A0150)
国家自然科学基金(61572032)
2018年度皖江高端装备制造协同创新中心开放基金(GCKJ2018009)
安徽省支持新能源汽车产业创新发展和推广应用项目资助