多因素优化蚁群算法机器人路径规划  

Colony Path Planning Based on Robot Ant Algorithm with Multi-Factor Optimization

作　　者：孙凌宇[1] 王威 刘文瀚 秦红亮 SUN Ling-Yu;WANG Wei;LIU Wen-Han;QIN Hong-Liang(School of Mechanical Engineering,Hebei University of Technology,Tianjin 300401,China)

机构地区：[1]河北工业大学机械工程学院,天津300401

出　　处：《计算机仿真》2024年第3期470-476,共7页Computer Simulation

基　　金：国家自然科学基金联合基金项目(U1913211);河北省应用基础研究计划重点基础研究项目(17961820D)。

摘　　要：采用传统蚁群算法进行路径规划中,存在收敛速度慢,路径不平滑,方向性与目的性较差等问题,提出了多因素优化蚁群算法以提高路径寻优的性能。应用扩散方式赋予地图不均匀的初始信息素,为路径搜索提供更好的方向性,避免了局部最优解的出现;距离启发信息、障碍物阻力启发信息和路径角度启发信息作为综合启发信息应用于蚁群算法,增强蚂蚁移动的目的性,缩短路径长度;运用贝塞尔曲线优化路径上的拐点处,输出平滑路径。仿真结果表明,多因素优化蚁群算法应对多样性环境具有更强的适应性,有效避免了局部最优解,规划出的最优路径更短且光滑,有利于机器人流畅通行,更具有工程实践意义。In the path planning using traditional ant colony algorithm,there are problems such as slow convergence speed,unsmooth path,and poor directionality and purpose.Therefore,a multi-factor optimization ant colony algorithm was proposed to improve the performance of path optimization.The diffusion method was used to endow the map with uneven initial pheromone,which can provide better directionality for path search and avoids the appearance of local optimal solutions.Distance heuristic information,obstacle resistance heuristic information and path angle heuristic information were utilized as comprehensive heuristic information,to enhance the purpose of ant movement and shorten the length of the path.The Bezier curve was used to optimize the inflection point on the path and output a smooth path,which provides a method basis for the efficient operation of the robot.The simulation results show that the multi-factor optimization ant colony algorithm has fast convergence speed,good stability,and the optimal path is shorter and smoother,which is more meaningful for engineering practice.

关 键 词：蚁群算法 路径规划 扩散 综合启发信息 贝塞尔曲线 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]