改进PSO及其在机器人路径规划中的应用  

Improved PSO and Its Application in Robot Path Planning

作　　者：王佳荣 周超 WANG Jia-rong;ZHOU Chao(Unit 92228 of the PLA,Beijing 100072,China)

机构地区：[1]中国人民解放军92228部队,北京100072

出　　处：《计算机仿真》2024年第5期436-440,445,共6页Computer Simulation

摘　　要：在解决多通路机器人路径规划问题时,标准粒子群优化算法容易早熟收敛陷入到局部最优通路中,且由于仅仅利用算法群体最优和个体最优的信息,最终搜索精度往往不高。提出一种粒子群优化算法的改进方案,针对算法寻优精度不高的问题,利用三角函数的单调性适时调整算法的惯性权重系数和加速系数,使得关键参数在迭代过程中保持最优匹配;针对算法容易陷入局部最优的问题,对较差适应度函数值对应的种群作变异操作,随机产生新的粒子来代替这部分较差粒子。最后通过简单和复杂两组路径规划对比实验,验证了改进算法在解决机器人路径规划问题时,具有寻优精度高、鲁棒性好的优点。When solving the multi-path robot path planning problem,standard particle swarm optimization algorithm is prone to precocious convergence into the local optimal path.In addition,the final search accuracy is often not high because only the information of the optimal group and the optimal individual is used.This paper presents an improved particle swarm optimization algorithm.Firstly,in order to improve the optimization accuracy of the algorithm,we adjusted the inertia weight coefficient and acceleration coefficient of the algorithm by using the monotonicity of trigonometric function.The key parameters can be matched optimally in the whole running of the algorithm.Secondly,in order to effectively avoid the algorithm falling into local optimum,the mutation operation was performed on the population corresponding to the poor fitness function value,and the new particles were randomly generated to replace this part of the poor particles.Finally,through the comparison experiments of simple and complex path planning,it was verified that the improved algorithm has the advantages of high accuracy and good robustness when solving the robot path planning problem.

关 键 词：粒子群优化算法 路径规划 参数调整 种群变异 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]