改进自适应精英蚁群算法的机器人路径规划  

Improved Adaptive Elite Ant Colony Algorithm for Robot Path Planning

作　　者：王影[1] 王晓茹 孙万龙 刘麒[1] WANG Ying;WANG Xiaoru;SUN Wanlong;LIU Qi(School of Information and Control Engineering,Jilin Institute of Chemical Technology,Jilin City 132022,China;Bosch Automotive Products(Changchun)Co.,Ltd,Changchun 130000,China)

机构地区：[1]吉林化工学院信息与控制工程学院,吉林吉林132022 [2]博世汽车部件(长春)有限公司,吉林长春130000 [3]吉林化工学院

出　　处：《吉林化工学院学报》2024年第3期1-8,共8页Journal of Jilin Institute of Chemical Technology

基　　金：吉林市科技局项目(201750244);吉林市科技局项目(20190502118);吉林化工学院科研项目(2018064);吉林化工学院重大科技项目(2016033);吉林化工学院重大科技项目(2018017)。

摘　　要：针对基本蚁群算法在移动机器人二维栅格地图存在路径规划时间长、收敛速度慢、迭代稳定次数多等问题,提出了一种改进的自适应精英蚁群算法。该算法通过引入距离参数因子改进启发式信息函数,采用自适应伪随机状态转移规则选择下一节点,同时融合角度引导因子到转移概率中以减少搜索盲目性,从而缩短搜索时间。此外,还定义了一种自适应信息素权重更新策略,仅对当代寻找到的最优路径进行信息素奖励,进一步提高了收敛速度。通过消融实验、不同规模和环境下的对比实验表明,改进后的算法规划的路径更优、收敛速度更快,验证了该算法的优越性和可行性。Aiming at the basic ant colony algorithm's problems of long path planning time,slow convergence speed,and high number of iterative stabilization in 2D grid maps for mobile robots,an improved adaptive elite ant colony algorithm was proposed.The algorithm improved the heuristic information function by introducing a distance parameter factor,selected the next node by using an adaptive pseudo-random state transfer rule,and also fused the angle guidance factor into the transfer probability to reduce the search blindness,thus shortening the search time.In addition,an adaptive pheromone weight updating strategy was defined to reward the pheromone only for the optimal paths found in the contemporary search,which further improved the convergence speed.The ablation experiments,comparative experiments under different scales and environments showed that the improved algorithm plans better paths and converges faster,verifying the superiority and feasibility of the algorithm.

关 键 词：自适应精英蚁群 路径规划 距离参数因子 角度引导因子 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]