基于改进遗传算法的AGV路径规划  

AGV optimal path planning based on improved genetic algorithm

作　　者：孙波 姜平[1] 周根荣[1] 董殿永[2] SUN Bo;JIANG Ping;ZHOU Gen-rong;DONG Dian-yong(School of Electrical Engineering,Nantong University,Nantong 226019,China;Library,Nantong University,Nantong 226019,China)

机构地区：[1]南通大学电气工程学院,江苏南通226019 [2]南通大学图书馆,江苏南通226019

出　　处：《计算机工程与设计》2020年第2期550-556,共7页Computer Engineering and Design

基　　金：南通市应用基础研究—工业创新基金项目（GY12017018）

摘　　要：为解决基本遗传算法在规划AGV运行路径时存在早熟收敛的问题,对基本遗传算法进行改进优化。用模拟退火法进行种群选择,提高种群的差异性;改进交叉、变异算子自整定策略和精英策略,提高算法的收敛速度;在适应度函数中加入路径曲折度、路径繁忙度和车辆负重度等多个规划指标,使规划出的路径更符合实际。将优化后的算法与基本遗传算法进行比较,仿真结果表明,改进后算法在AGV路径规划中具有高效性。To solve the problem that the basic genetic algorithm has premature convergence when planning the AGV running path,the basic genetic algorithm was improved and optimized.The population selection was improved by the idea of simulated annealing,which improved the diversity of the population.The crossover and mutation operator self-tuning strategies and elite strategies were improved,and the convergence speed of the algorithm was improved.Multiple planning indicators such as path tortuosity,path busyness and vehicle load-bearing were added to the fitness function to make the planned path more realistic.The optimized algorithm was compared with the basic genetic algorithm.Results of simulation show that the improved algorithm is efficient in AGV path planning.

关 键 词：遗传算法 路径规划 模拟退火 自整定策略 精英策略 多指标规划 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构] TP242[自动化与计算机技术—计算机科学与技术]