基于遗传蚁群动态融合的地面自主作战机器人路径规划  

A Dynamic Fusion Algorithm of Path Planning Based on Genetic and Ant Colony for Ground Autonomous Combat Robot

作　　者：李郁峰 李魁武[2] 潘玉田 郭保全 余红英 LI Yufeng;LI Kuiwu;PAN Yutian;GUO Baoquan;YU Hongying(School of Electrical and Control Engineering,North University of China,Taiyuan 030051,Shanxi,China;Northwest Institute of Mechanical&Electrical Engineering,Xianyang 712099,Shaanxi,China)

机构地区：[1]中北大学电气与控制工程学院,山西太原030051 [2]西北机电工程研究所,陕西咸阳712099

出　　处：《火炮发射与控制学报》2019年第4期42-46,50,共6页Journal of Gun Launch & Control

基　　金：山西省应用基础研究计划项目(201601D102029)

摘　　要：在机器人路径规划与避障算法中,遗传算法具有快速全局搜索能力,但是没有利用系统中反馈的信息。蚁群算法具有很好的信息反馈性,但是由于初期信息素匮乏导致求解速度较慢,易陷入局部最优。提出了一种动态融合的方法,在算法初期通过遗传算法生成蚁群算法的初始信息素分布,后期采取蚁群算法动态融合遗传算子的方法。通过路径规划仿真及实验分析,该动态融合算法不仅提高了收敛速度,而且改善了蚁群算法易陷入局部最优的问题;同时引入了动态避障策略,从而达到了更好的路径规划效果。In robot path planning and obstacle avoidance algorithm,genetic algorithm has the ability of fast global search,but it does not make use of the feedback information in the system.Ant colony algorithm has a good information feedback,but due to the scarcity of pheromones in the early stage,the solution speed is slow and it is prone to fall into the local optimum.In this paper,a dynamic fusion method was proposed to generate the initial pheromone distribution of ant colony algorithm by genetic algorithm in the early stage of the algorithm,with ant colony algorithm adopted in the later stage to dynamically fuse genetic operators.Through path planning simulation and experimental analysis,the convergence speed was improved,with the problem of ant colony algorithm being easy to fall into local optimum made less serious in the dynamic fusion algorithm.At the same time,dynamic obstacle avoi-dance strategy was introduced to achieve better path planning effects.

关 键 词：路径规划 遗传算法 蚁群算法 动态融合 

分 类 号：TJ303+.3[兵器科学与技术—火炮、自动武器与弹药工程] TP242[自动化与计算机技术—检测技术与自动化装置]