加入动态搜索模型的蚁群算法及其应用  

Ant colony algorithm with dynamic search model and its application

作　　者：马晓平[1] 赵学涛 王炬成[1] MA Xiao-ping;ZHAO Xue-tao;WANG Ju-cheng(College of Naval Architecture and Ocean Engineering,Jiangsu University of Science and Technology,Zhenjiang 212000,China)

机构地区：[1]江苏科技大学船舶与海洋工程学院,江苏镇江212000

出　　处：《计算机工程与设计》2023年第11期3462-3468,共7页Computer Engineering and Design

基　　金：工业和信息化部高技术船舶基金项目(MC-201917-C09)。

摘　　要：用传统蚁群算法求解复杂条件下的路径规划问题时易出现局部最优、收敛速度慢等问题。提出利用栅格法建立地图模型,加入动态搜索模型对蚁群算法进行优化,包括根据不同的初始、终止位置对路径选择进行限制;对死锁法要删除的栅格进行优化;在信息素浓度更新过程中,设置动态阀值,对要增加信息素浓度但长度低于阀值的路径进行“惩罚”,设置信息素浓度下限。通过对路径规划与TSP问题等实例的仿真,改进蚁群算法在避免局部最优、加快收敛速度方面均优于对比算法,验证了改进算法在提高收敛效率、缩短计算时间等方面具有优越性。When using the traditional ant colony algorithm to solve the path planning problem under complex conditions,it is easy to have problems such as local optimization and low convergence speed.A map model based on grid method was proposed,the ant colony algorithm was optimized by adding dynamic search model,including restrict path selection according to different initial and termination locations.The deadlock method was optimized to remove the grid.In the process of updating pheromone concentration,dynamic threshold was set,the path with a length below the threshold where the pheromone concentration needed to be increased was punished,at the same time,the lowest limit of the pheromone concentration was set.Through the simulation of path planning and TSP problems,the improved ant colony algorithm is better than the comparison algorithm in avoiding local optimization and accelerating convergence speed.The improved algorithm shows advantages in improving convergence efficiency and shortening calculation time.

关 键 词：栅格地图 蚁群算法 路径规划 动态搜索模型 死锁法 动态阀值 信息素浓度更新 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]