摘要
【目的】针对传统蚁群算法(ant colonyalgorithm, ACA)在移动机器人(automatic guided vehicle, AGV)路径规划中搜索效率低、寻找路径长、拐点个数多等问题,提出一种改进的蚁群优化算法(ant colony optimization, ACO)。【方法】首先,在蚁群算法中加入预估代价值策略来改进启发函数,增强目标点的引导作用,提升搜索效率;然后,结合狼群算法(wolf pack algorithm, WPA)分配机制来更新信息素,解决路径规划时易陷入局部最优的问题;接着加入拐点影响因子来降低路径拐点;最后,采用动态避障策略来解决死锁问题。【结果】运用改进蚁群优化算法后,移动机器人路径规划时,最佳路径长度、迭代次数和拐点数等比传统算法分别降低9.7%、57.8%、65.0%。【结论】本研究结果能为移动机器人在复杂环境下的路径选择提供重要参考。
[Objective]Aiming at the problems of low search efficiency,long search path,and multiple inflection points in traditional ant colony algorithm(ACA)for automatic guided vehicle(AGV)path planning,an improved ant colony optimization(ACO)algorithm was proposed.[Method]First,an estimated surrogate value strategy was incorporated into the ant colony algorithm to improve the heuristic function,enhance the guiding effect of the target point,and boost search efficiency;then,the wolf pack algorithm(WPA)allocation mechanism was combined to update pheromones,solving the problem of easily falling into local optima during path planning,and adding inflection point influence factors to reduce path inflection points;finally,a dynamic obstacle avoidance strategy was adopted to solve the deadlock problem.[Result]After applying the improved ant colony optimization algorithm,the optimal path length,the number of iterations and the number of inflections in AGV path planning are reduced by 9.7%,57.8%,and 65.0%,respectively,compared to traditional algorithms,[Conclusion]This study provides important references for AGV to choose paths under complex environmental conditions.
作者
陈岁繁
王浈元
李其朋
CHEN Suifan;WANG Zhenyuan;LI Qipeng(School of Mechanical and Energy Engineering,Zhejiang University ofScience and Technology,Hangzhou 310023,Zhejiang,China)
出处
《浙江科技学院学报》
CAS
2024年第1期59-67,共9页
Journal of Zhejiang University of Science and Technology
基金
浙江省科技计划项目(2023C02008,2024C04037)。
