改进蚁群算法的机器人路径规划  

Robot path planning based on improved ant colony optimization

作　　者：薛翔 朱其新 朱永红 XUE Xiang;ZHU Qixin;ZHU Yonghong(School of Electronic and Information Engineering,Suzhou University of Science and Technology,Suzhou 215009,Jiangsu,China;School of Mechanical Engineering,Suzhou University of Science and Technology,Suzhou 215009,Jiangsu,China;Jiangsu Intelligent Inclusion Robot Engineering Technology Center,Suzhou 215009,Jiangsu,China;Suzhou Key Laboratory of Eutectic Robot Technology,Suzhou 215009,Jiangsu,China;School of Mechanical and Electronic Engineering,Jingdezhen Ceramic University,Jingdezhen 333001,Jiangxi,China)

机构地区：[1]苏州科技大学电子与信息工程学院,江苏苏州215009 [2]苏州科技大学机械工程学院,江苏苏州215009 [3]江苏省智能共融机器人工程技术中心,江苏苏州215009 [4]苏州市共融机器人重点实验室,江苏苏州215009 [5]景德镇陶瓷大学机电工程学院,江西景德镇333001

出　　处：《西安工程大学学报》2024年第6期59-66,共8页Journal of Xi’an Polytechnic University

基　　金：国家自然科学基金资助项目(62063010);泰州市科技支撑计划资助项目(TG202117)。

摘　　要：针对蚁群算法在求解机器人路径规划时存在收敛速度慢、容易陷入局部最优等问题,提出了一种改进的蚁群算法。首先,建立了一种趋向启发函数,使得待选节点更趋于起点和终点的连线,对于避免局部最优起到一定的作用,在此基础上引入柯西分布函数,不断削弱趋向启发函数的影响效果,提高了算法后期的全局搜索能力;其次,改进了距离启发函数,综合考虑待选节点和起点以及待选节点和终点之间的距离关系,加快了算法的收敛速度;再次,提出了一种根据迭代次数动态调整的信息素挥发因子,不断减小信息素挥发因子直至合适的大小,增强了全局搜索能力;最后,采用三次B样条曲线进行路径平滑处理,平滑了路径,缩短了路径长度。仿真结果表明:改进后的算法相比传统算法,收敛时间减小了3%,最短路径长度缩短了12%,收敛迭代次数减少了76%。改进后的算法较传统算法最小路径长度更短,收敛速度更快,路径也更加平滑,证明了改进后的算法在解决收敛速度慢、容易陷入局部最优等问题上的有效性。Aimed at the problems of ant colony algorithm in solving robot path planning,such as slow convergence speed and tending to fall into local optimization,an improved ant colony optimization was proposed.Firstly,a trend heuristic function was established,makeing the nodes to be selected closer to the line between the starting point and the end point,which plays a certain role in avoiding the local optimization,and the Cauchy distribution function was introduced on the basis of this function,which constantly weakens the influence of the trend heuristic function,and improves the global search ability of the algorithm in the later stage.Secondly,the distance heuristic function was improved to speed up the convergence of the algorithm by integrating the distance relationship between the nodes to be selected and the starting point as well as between the nodes to be selected and the end point.Then,a pheromone volatilization factor that is dynamically adjusted according to the number of iterations was proposed,and the pheromone volatilization factor is continuously reduced until the appropriate size,which enhances the global search ability.Finally,a cubic B-spline curve was used for the path smoothing process,which smoothes the paths and shortens the length of the paths.Simulation results show that compared with the traditional algorithm,the improved algorithm reduces the convergence time by 3%,the shortest path length by 12%,and the number of convergence iterations by 76%.The improved algorithm has a shorter minimum path length,faster convergence speed and smoother paths than the traditional algorithm;it proves the effectiveness of the improved algorithm in solving the problems of slow convergence speed and easy to fall into local optimization.

关 键 词：机器人 路径规划 蚁群算法 启发函数 三次B样条曲线 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]