1 School of Mechanical and Electronic Engineering, Wuhan University of Technology, Wuhan 430070, China
2 SAIC-GM-Wuling Intelligent Manufacturing Department, Liuzhou 545007, China
Show less
文章历史+
Received
Published
Revised
2025-11-27
2026-10-28
2026-04-21
Online Date
2026-11-10
PDF (10279K)
摘要
针对复杂栅格环境下路径规划中传统算法易陷入局部最优、收敛速度慢及路径安全性不足等问题,提出了一种融合灰狼优化算法(grey wolf optimizer, GWO)与蚁群算法(ant colony optimization, ACO)的路径规划方法。该方法以栅格地图为环境模型,通过GWO优化ACO的关键参数,同时引入GWO中领导者的优秀路径信息增强ACO的信息素更新机制,该算法的核心亮点在于实现了宏观参数优化与微观信息素引导的深度闭环。为提升算法性能,对GWO进行改进:采用Logistic混沌映射初始化种群以提高多样性,设计自适应收敛因子和动态权重调整策略优化位置更新;通过函数约束障碍物附近的不安全对角移动,确保路径合法性。实验结果表明,与传统ACO或GWO相比,所提算法在大型栅格地图中能更快收敛至更短路径,且有效避开障碍物及不安全移动区域,验证了其在复杂环境下路径规划的有效性和优越性。该方法可为移动机器人、自动引导车等领域的自主导航提供技术支持。
Abstract
In order to address the issues of traditional algorithms in path planning under complex grid environments, such as easy entrapment in local optima, slow convergence speed, and insufficient path safety, a path planning method fusing the GWO (grey wolf optimizer) and ACO (ant colony optimization) was proposed. The grid map was taken as the environment model. Key parameters of ACO were optimized by GWO. Simultaneously, excellent path information from GWO leaders was introduced to enhance the pheromone update mechanism of ACO. The core highlight of this algorithm lies in the realization of a deep closed-loop between macroscopic parameter optimization and microscopic pheromone guidance. To enhance algorithm performance, the GWO was improved. A logistic chaotic map was adopted for population initialization to improve diversity. An adaptive convergence factor and a dynamic weight adjustment strategy were designed to optimize position updates. Furthermore, unsafe diagonal movements near obstacles were constrained by a function to ensure path legitimacy. The experimental results indicate that compared with traditional ACO or GWO, the proposed algorithm converges to shorter paths more quickly in large-scale grid maps. Obstacles and unsafe areas are effectively avoided. The effectiveness and superiority of the method in complex environment path planning are verified. This method can provide technical support for autonomous navigation of mobile robots and AGVs (automated guided vehicles).
YAN Qi-long, YU Xian-tao, ZHANG Hong.
Adaptive Ant Colony Path Planning Based on Wolf Pack Hierarchy Pheromone Feedback Mechanism[J].
Science Technology and Engineering, 2026, 26(30): 13110-13121 DOI:10.12404/j.issn.1671-1815.2508178
近年来,随着“中国制造2025”战略的深入实施与智慧物流产业的蓬勃发展,中国在自动导引车(automated guided vehicle,AGV)、自主移动机器人(autonomous mobile robot,AMR)等高端智能装备领域的市场规模与技术应用已跃居全球前列。在黑灯工厂、无人仓储等典型的高并发、高密度应用场景中,为移动机器人规划兼顾安全、高效与低能耗的无碰撞路径,已成为制约中国柔性制造系统跨越式发展的关键瓶颈。路径规划作为移动机器人自主导航的核心课题,其目标是在存在复杂障碍物的环境中,寻找一条从起点到终点的最优或次优安全路径。特别是在复杂仓储环境下,工业AGV的频繁急转弯会急剧增加底盘电机的机械磨损与运行能耗。因此,近年来国内学者高度聚焦于AGV路径的平滑度优化。例如,杨桂华等[1]针对仓储环境探讨了减少移动机器人路径转折次数的策略;袁秀怡等[2]以综合能耗最低为导向,提出了消除冗余拐点的优化方法;张志清等[3]则利用Logistic混沌映射改进算法,大幅提升了实际物流配送路径的求解效率;郑亮等[4]则针对工业AGV转弯过多的痛点,引入平滑机制以更贴合实际应用需求。这些研究充分表明,在追求路径最短的同时,提升路径平滑度已成为当前智能导航领域的刚性需求。路径规划主要包括3个部分:地图模型的建立、初始路径的探索、优化路径的生成。二维地图模型通常通过栅格地图、拓扑地图等来进行建立。近年来,基于栅格地图的建模方法因其直观性和易实现性被广泛应用。在此背景下,各类智能优化算法,如D*算法[5]、A*算法[6]、蚁群算法(ant colony optimization,ACO)以及粒子群算法(particle swarm optimization, PSO)[7]等,被成功应用于路径规划问题。
其中,蚁群算法模拟蚂蚁群体的觅食行为,通过信息素的正反馈机制寻找最优路径,因其强大的并行搜索能力和鲁棒性而备受关注。但是传统蚁群算法存在明显的局限性,如性能高度依赖于参数设置,参数选取不当极易导致收敛速度慢或陷入局部最优。由于在搜索起始阶段缺乏足够的信息素积累,传统蚁群算法往往陷入盲目搜索,从而制约了整体规划效率。针对蚁群算法的这种固有弊端,学术界对此进行了广泛的改进研究。葛志远等[8]为了突破算法耗时长及搜索效能低下的瓶颈,通过重构启发因子,并引入优胜劣汰规则与全局信息素动态调节机制,有效提升了寻径效率并缓解了程序死锁现象;李开荣等[9]聚焦于提升路径的平滑度,为解决传统算法中转弯角度过大的缺陷,提出了一种带有角度约束的优化算法。该方法不仅强化了初始信息素的导向作用,还融合狼群分配机制与MMAS(max-min ant system)策略,从而显著提高了路径质量;余稼洋等[10]提出改进元胞蚁群算法,利用势场修正启发函数,采用差别搜索策略与自适应信息素更新方式,有效提升路径规划速度与全局搜索能力;邓冬冬等[11]融合改进蚁群算法与人工势场法,优化信息素更新规则与启发信息,改进斥力势场函数并设置子目标点,增强复杂环境中路径规划性能;孟文俊等[12]创新性地提出了一种双层级寻路机制,旨在显著提升算法的收敛效率;金将等[13]为提高蚁群算法初始信息素分布的导向作用,提出基于A*算法生成初始路径并非均匀分配信息素,结合转角与路径匝数优化信息素更新,并融合动态窗口法增强局部避障能力。目前的学术研究更倾向于采用多算法融合的手段来突破单一策略的性能瓶颈。例如,Fu等[14]利用双向人工势场法辅助蚁群进行搜索优化;Sheng等[15]则将鹅优化算法与蚁群算法相结合。此外,针对传统ACO在复杂环境中易陷入死锁且搜索盲目的问题,武丁杰等[16]通过引入启发式信息分配初始信息素,有效加速了算法的收敛;杨玉等[17]通过引入惩罚信息素有效防止了智能体的无效搜索;张浩等[18]则通过提取路径特征参数动态调整启发因子,降低了算法前期的盲目性。此外,针对自动导引车(AGV)的路径规划难题,车健波等[19]动态融合蚁群与遗传算法,通过算法自适应信息素分布与挥发因子调节,实现全局与局部搜索的互补优化。
在针对蚁群算法的现有研究中,大多聚焦于内部框架的修正(如优化启发式函数、更新信息素规则等)以及同级算法的融合,但蚁群算法的性能严重依赖于参数(如信息素因子、启发式因子等)的设定,而现有方法多依赖于经验试错,缺乏一种高效、自适应的参数优化机制。与此同时,灰狼优化算法(grey wolf optimizer, GWO)因其卓越的全局搜索能力,逐渐被引入路径规划领域。孙波等[20]通过引入非线性控制参数与动态扰动,显著提升了GWO的收敛精度与避障效果。朱喜明等[21]通过改进GWO的初始化方式与收敛因子,成功解决了移动机器人在复杂地形下的寻优难题。这些前沿研究为本文优化算法参数提供了坚实的理论依据。然而,标准的GWO原生设计用于连续空间参数寻优,若直接应用于离散的栅格环境,极易因缺乏有效的节点转移机制而生成不可行路径。
YangGuihua, WeiJiale. Path planning of logistics robots based on improved A* and DWA algorithms[J]. Science Technology and Engineering, 2022, 22(34): 15213-15220.
YuanXiuyi, YuZhen, WangDong. Path planning for AGV handling in complex warehouses based on improved ant colony algorithm with minimum energy consumption[J]. Science Technology and Engineering, 2025, 25(32): 13873-13882.
ZhangZhiqing, WangShuo, ZhuLeipeng, et al. Path planning and application of logistics terminal distribution based on improved grey wolf optimization algorithm[J]. Science Technology and Engineering, 2025, 25(34): 14890-14899.
ZhengLiang, SunLonglong, ChenShuang. An improved path planning algorithm for industrial automated guided vehicles[J]. Science Technology and Engineering, 2021, 21(16): 6758-6763.
[9]
IvanM, MarijaS, KrunoL, et al. Path planning for active SLAM based on the D* algorithm with negative edge weights[J]. IEEE Transactions on Systems, Man, and Cybernetics: Systems, 2018, 48(8): 1321-1331.
[10]
TangX, ZhuY, JiangX. Improved A-star algorithm for robot path planning in static environment[J]. Journal of Physics: Conference Series, 2021, 1792(1): 012067.
[11]
YoukyungH, SuseongK, JihunC. Integrated global and local path planning for quadrotor using particle swarm optimization[J]. IFAC PapersOnLine, 2020, 53(2): 15621-15625.
Li KairongLiuShuang, HuQianqian, et al. Path planning based on improved ant colony optimization algorithm with turning angle constraint[J]. Journal of Computer Applications, 2021, 41(9): 2560-2568.
DengDongdong, XuJianmin, MengHan, et al. Mobile robot path planning based on fusion of ant colony algorithm and artificial potential field method[J]. Chinese Journal of Scientific Instrument, 2025, 46(2): 1-16.
MengWenjun, ChengWenlong, WangYijin, et al. AGV path planning based on two-level system ant colony algorithm[J]. Machinery Design & Manufacture, 2026(2): 1-7.
JinJiang, WangXiaoping, ZangTiegang, et al. Robot obstacle avoidance path planning based on improved ant colony algorithm[J]. Computer Engineering and Design, 2025, 46(4): 950-958.
[24]
FuB, ChenY, QuanY, et al. Bidirectional artificial potential field-based ant colony optimization for robot path planning[J]. Robotics and Autonomous Systems, 2025, 183: 104834.
[25]
ShengX, YangJ, YouL, et al. GOA-ACO: agoose optimized ant colony algorithm for the automated guided vehicle path planning[J]. Alexandria Engineering Journal, 2025, 130: 724-737.
WuDingjie, LiuGerui, XieWeizheng. Rerouting path planning based on A*-improved ant colony algorithm[J]. Science Techno-logy and Engineering, 2025, 25(32): 14053-14061.
YangYu, JiangXiaowei, ChenRuotong, et al. Ship path planning based on penalty pheromone ant colony algorithm[J]. Scie-nce Technology and Engineering, 2025, 25(10): 4246-4255.
CheJianbo, TangDonglin, HeYuanyuan, et al. Research on dynamic path optimization method fusing ant colony and genetic algorithm[J]. Chinese Journal of Engineering Design, 2025, 32(6): 789-802.
SunBo, ZhouJiankang, ZhaoYuqing, et al. Global path planning of robots based on improved grey wolf optimization algorithm[J]. Science Technology and Engineering, 2024, 24(33): 14287-14297.
ZhuXiming, NiHengxin, ZhaoJianpeng. Three-dimensional path planning of mobile robots based on improved grey wolf algorithm[J]. Science Technology and Engineering, 2025, 25(3): 1125-1132.
[38]
LiJ, TangW, ZhangD, et al. Map construction and path planning method for mobile robots based on collision probability model[J]. Symmetry, 2023, 15(10): 1891.
[39]
TarissanF, QuoitinB, MérindolP, et al. Towards a bipartite graph modeling of the internet topology[J]. Computer Networks, 2013, 57(11): 2331-2347.
[40]
MirjaliliS, MirjaliliM S, LewisA. Grey wolf optimizer[J]. Advances in Engineering Software, 2014, 69: 46-61.
YangYiying, PanDazhi. Improved grey wolf optimizationalgorithm based on mutation operator and chaotic mapping[J]. Intelligent Computer and Applications, 2025, 15(6): 58-66.