基于改进K-means聚类和量子粒子群算法的多航迹规划

针对在复杂环境下需要通过多航迹规划以实现武器协同的问题,利用排挤机制产生Kmeans聚类的初始聚类中心,并将改进K-means聚类与量子粒子群算法(QPSO)相结合应用于无人机的三维多航迹规划。改进算法解决了K-means聚类易陷入局部最优、聚类准确率低的问题。根据产生的初始聚类中心,将粒子划分成多个子种群,利用QPSO算法对每个子种群进行优化,使得每个子种群可以产生一条可行航迹。仿真分析证明了改进算法可以有效保证子种群之间的多样性,生成较为分散的多条可行航迹。

Multiple route planning based on improved K-means clustering and quantum-behaved particle swarm optimization

For the problem of multiple routes planning to realize the weapon cooperation in complex environment,K-means clustering is improved by an exclusion mechanism which generates the initial cluster centers.A method combining quantum-behaved particle swarm optimization(QPSO) with K-means clustering is proposed and applied to 3-D multiple routes planning of unmanned aerial vehicle(UAV).The improved algorithm solves the problem of falling in local best and improves the clustering accuracy.It classifies the particles to several subgroups.Then every subgroup is optimized by QPSO so as to generate a feasible route.Finally,multiple and dispersive routes are constituted.Simulation proves that the improved algorithm can assure the variety of subgroups and generates feasible and diverse routes.