摘要
针对模糊C均值聚类航迹关联算法存在局部最优问题,同时算法的收敛速度受初始值的影响也较大,提出一种将粒子群(PSO)和模糊C均值(FCM)聚类算法相结合的航迹关联算法。该算法将多目标的航迹关联问题看做是一类约束条件下的组合优化问题,利用粒子群(PSO)强大的全局寻优能力,与模糊C均值聚类算法相结合求解航迹关联问题。仿真结果表明:在相同的条件下,粒子群优化的模糊C均值聚类算法与模糊C均值聚类算法相比,聚类性能明显改善,关联正确率也有明显的提高。
To resolve the problems of local optima and slow convergence of the fuzzy C-means algorithm,an clustering algorithm of track fusion combing particle swarm optimization with fuzzy C-means was proposed.Track association was regarded as a kind of constrained combination optimization problem,which could be solved by particle swarm optimization fuzzy C-means algorithm.Particle swarm optimization fuzzy C-means algorithm and fuzzy C-means algorithm were respectively adopted to solve track association.Simulation results indicated that the particle swarm optimization fuzzy C-means algorithm could improve the clustering performance and association accuracy.
出处
《探测与控制学报》
CSCD
北大核心
2010年第6期26-29,共4页
Journal of Detection & Control
关键词
目标跟踪
航迹关联
PSO
FCM
target tracking
track-association
particle swarm optimization
fuzzy C-means