农业机械学报2024，Vol.55Issue(2)：423-432,449,11.DOI:10.6041/j.issn.1000-1298.2024.02.042
基于动态扩展邻域蚁群算法的移动机器人路径规划
Path Planning of Mobile Robots Based on Dynamic Extended Neighbourhoods Ant Colony Algorithm
摘要
Abstract
To solve the problems of ant colony algorithm in complex grid environment,such as local optimization,many turning points and slow convergence,dynamic extended neighbourhoods ant colony optimization(DENACO)algorithm was proposed.Firstly,the method of dynamic extended neighborhoods was applied in the ant search mode to obtain the optimal convergence path length and reduce the number of inflection points and the number of path nodes.Meanwhile,a computational method and increment rule of pheromone were defined to reduce space costs,and the upper and lower limits of pheromone were set to avoid premature convergence of the algorithm to local optimality.Secondly,the adaptive adjustment factor and target point factor were introduced into the heuristic function,and a weight coefficient was set to improve the global search ability of the algorithm.Moreover,an iteration threshold of the algorithm was set.When the iteration exceeded the threshold,the pheromone concentration factor and heuristic factor values were updated to improve the convergence speed of the algorithm.Finally,a double optimal strategy of nodes of path was proposed.Two optimization methods were used to further optimize the planned path,and the best was taken as the final optimization result to improve the comprehensive quality of the path.Simulation experiments on raster maps of different complexities and scales showed that compared with the traditional ant colony algorithm and other comparison algorithms,the path planned by DENACO algorithm was superior.It had a shorter path length,reduced number of inflection points,accelerated convergence speed,and significantly fewer path nodes.These results indicated that the DENACO algorithm was highly feasible and applicable.关键词
移动机器人/蚁群算法/路径规划/动态扩展邻域/自适应启发函数Key words
mobile robot/ant colony algorithm/path planning/dynamic extended neighbourhoods/adaptive heuristic function分类
信息技术与安全科学引用本文复制引用
潘玉恒,奥日格拉,鲁维佳,丛佳,王世通,陈阳..基于动态扩展邻域蚁群算法的移动机器人路径规划[J].农业机械学报,2024,55(2):423-432,449,11.基金项目
国家自然科学基金项目(62204168)、天津市科技计划项目(20YDTPJC00160、21YDTPJC00780)、天津市教委科研计划项目(2019KJ101、2017SK027)、天津市研究生科研创新项目(2022SKYZ033)和天津城建大学教育教学改革与研究重点项目(JG-ZD-22035、JG-ZD-22038) （62204168）
