苑光明 翟云飞 丁承君 张鹏

[摘要]针对AGV在自动化生产线中原有路径规划算法存在路径拐弯次数多，不利于AGV自动控制的问题，提出了一种改进遗传算法。为提高AGV运行的效率，该算法引入了拐弯因素。针对在路径规划中传统遗传算法收敛速度慢的问题，结合分层方法，改进传统的精英保留策略。在算法进化过程中，根据个体适应度的变化动态调整交叉概率和变异概率，加快算法的收敛速度。Matlab仿真实验结果显示：改进遗传算法能够规划出一条更合理的路径，相比较传统方法减少了转弯次数，改善了搜索路径质量，表明该算法可以满足自动化生产线AGV路径规划的要求。

[关键词]自动导航车；路径规划；改进遗传算法；转弯次数

[中图分类号]TP 273[文献标志码]A[文章编号]10050310（2018）01006505

AGV Path Planning Based on Improved Genetic Algorithm

Yuan Guangming， Zhai Yunfei， Ding Chengjun， Zhang Peng

（Institute of Mechanical Engineering，Hebei University of Technology，

Tianjin 300132，China）

Abstract： In order to solve the problem that the path planning algorithm in AGV automation production line has the number of turns， which is not conducive to the automatic control of AGV， an improved genetic algorithm is proposed. In order to improve the efficiency of AGV automatic control， the algorithm introduces the turning factor. Aiming at the problem of slow convergence of the path planning in the traditional genetic algorithm， the traditional elitism strategy is improved with hierarchical method. In the process of evolutionary algorithm， the crossover probability and mutation probability are dynamically adjusted according to the change of individual fitness， and the convergence speed of the algorithm is accelerated. Matlab simulation results show that the improved genetic algorithm can plan a more reasonable path， and the number of turns is reduced compared with the traditional methods， and the quality of the search path has been improved， which show that the algorithm can meet the requirements of automated production line AGV path planning.

Keywords：

Automated Guided Vehicle（AGV）； Path planning； Improved genetic algorithm； Turn times

0引言

隨着自动化技术的不断发展，目前国内大部分制造业，尤其是在汽车制造、制药等劳动力密集的制造企业，传统的物料运输方式效率低，柔性较差，且需要的人工量大，对于企业来说难以达到其高效生产的要求。为了克服这种现状，相关领域积极引入AGV（Automated Guided Vehicle）自动导航车，达到物料运输的目的[12]。AGV在实际应用中仍然有一些需要解决的问题，路径规划是其中比较重要的一个问题，当AGV收到调度系统下达的任务后，会自动规划出1条从当前位置到达目标位置的路径，该路径需要优化的方面有行程时间、行程距离、所需能耗等[3]。AGV路径规划可以抽象建模成多目标优化问题，多目标通过惩罚函数使其成为单目标优化问题[4]。……