基于DHPA^(*)-DSACO算法的AGV路径规划研究  

Research on AGV Path Planning Based on DHPA^(*)-DSACO Algorithm

作　　者：王俊岭[1] 刘佳年 边俊君 王振东 WANG Junling;LIU Jianian;BIAN Junjun;WANG Zhendong(School of Information Engineering,Jiangxi University of Science and Technology,Ganzhou Jiangxi 341000,China)

机构地区：[1]江西理工大学信息工程学院,江西赣州341000

出　　处：《机床与液压》2025年第5期15-23,共9页Machine Tool & Hydraulics

基　　金：国家自然科学基金地区科学基金项目(62062037);江西省自然科学基金项目(20212BAB202014)。

摘　　要：自主引导车(AGV)的路径规划算法是确保其正常运行的关键部分。针对A^(*)算法在路径规划过程中存在的搜索效率低、路径曲率大的问题,以及蚁群ACO算法收敛速度慢和对参数敏感等缺陷,提出一种动态启发式惩罚A^(*)与动态感知蚁群优化算法相融合的算法—DHPA^(*)-DSACO。DHPA^(*)算法通过设置动态权重因子,结合父节点启发距离,并引入转弯惩罚项,以降低运行时间和路径曲率。DSACO算法通过设置自适应蚁群启发因子和动态挥发因子,优化信息素更新策略,从而缩短路径长度。同时,该算法利用B样条曲线对路径进行平滑处理。为验证算法的可行性,在PyCharm环境中将DHPA^(*)-DSACO算法与其他算法进行对比测试,并对实验结果进行了分析。最后,为了模拟真实世界中的情况,基于ROS系统建立仿真平台,验证了DHPA^(*)-DSACO算法的有效性。结果表明:DHPA^(*)-DSACO算法有效降低了路径长度、曲率和运行时间,显著提升了运行效率。此外,该算法还能有效避免算法陷入局部最优解,减少收敛迭代次数,进一步增强了算法的鲁棒性,使其更好地适应AGV的实际运行情况。The path planning algorithm of an autonomous guided vehicle(AGV)is a critical part of ensuring its proper operation.Aiming at the problems of A^(*)algorithm in path planning such as low search efficiency,large path curvature,slow convergence and sensitivity to parameters of ACO algorithm,a dynamic heuristic punitive A^(*)(DHPA^(*))and dynamic sensing ant colony optimization(DSACO)fusion algorithm was proposed.The DHPA^(*)algorithm was developed to mitigate high path curvature by incorporating dynamic weighting factors,considering the heuristic distance from parent nodes,and introducing penalties for turns.Meanwhile,DSACO was optimized with adaptive ant colony heuristic factors and dynamic evaporation factors to enhance the pheromone update strategy.B-spline curves were applied to smooth the paths.Experimental tests were conducted in the PyCharm environment,and simulations on a ROS-based platform were performed to verify the effectiveness of the DHPA^(*)-DSACO algorithm.The results show that the DHPA^(*)-DSACO algorithm effectively reduces path length,curvature,and runtime,significantly improving operational efficiency.Furthermore,it prevents the algorithm from being trapped in local optima,reduces the number of convergence iterations,and enhances robustness,making it highly suitable for real-world AGV operations.

关 键 词：路径规划 蚁群算法 A^(*)算法 B样条曲线 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]