摘要
粒子群优化(PSO)粒子滤波算法容易陷入局部最优,从而降低算法精度。针对该问题,提出一种基于混沌的PSO粒子滤波算法。该算法通过混沌搜索算法找到全局最优位置,驱散聚集在局部最优的粒子群,使其向全局最优位置靠近,增加有效估计粒子数,抑制粒子退化与枯竭问题。仿真结果表明,与传统的粒子滤波算法和PSO粒子滤波算法相比,改进算法的估计精度有较大提高。
Particle Swarm Optimization Particle Filtering（PSOPF） algorithm is easy to fall into local optimum, so the particles can not move to the global optimal location, and reduce algorithm precision. According to this problem, the paper proposes a Particle Swarm Optimization Particle Filtering based on Chaotic（CPSOPF） algorithm. Through the chaotic search algorithm, this algorithm makes particles find the global optimal location, dispels particle swarm at local optimum location and makes them move to global optimal location. So the number of effective particles increases, which can effectively restrain particles degradation and exhaustion. Simulation results show that the CPSOPF algorithm can remarkably improve the estimation accuracy compared with of the conventional Particle Filtering（PF） and the traditional PSOPF algorithm.
出处
《计算机工程》
CAS
CSCD
2012年第8期134-136,140,共4页
Computer Engineering
基金
甘肃省自然科学基金资助项目(1014RJZA028)
关键词
粒子滤波
混沌搜索算法
粒子群优化算法
局部最优
粒子退化
粒子枯竭
Particle Filtering（PF）
chaotic search algorithm
Particle Swarm Optimization（PSO） algorithm
local optimal
particle degeneracy
