Particle Filter Based on Memetic Algorithm for Nonlinear System State Estimation
A new method based on memetic algorithm (MA) to optimize the particle filter (PF) is proposed in this paper.The particle filter is typically crucial to deal with the non-linear,non-Gaussian problems.It has an inherent sample degeneracy problem.Resampling methods are solving the problem.However,the resampling stages cause a new problem,which is the failure to maintain the diversity of particles,called sample impoverishment.In order to increase the diversity of particles,we introduce memetic algorithm into the particle filter process,a hybrid filter method called memetic algorithm based particle filter (MAPF).In the particle filter,memetic algorithm combines the global search with the local search to increase the diversity of particles and the number of meaningful particles.The experimental results of non-linear system show that the performance of MAPF is better than several variants of particle filter with different resampling algorithms.
Particle filter Sample impoverishment Memetic algorithm
Qin Liu Lei Yuan Zhihua Cai Qiong Gu
School of Mathematics and Computer Science,Hubei University of Arts and Science,Xiangyang Hubei 4410 School of Mathematics and Computer Science,Hubei University of Arts and Science,Xiangyang Hubei 4410 School of Computer Science,China University of Geosciences(Wuhan),Wuhan Hubei 430074,China Institute of Logic and intelligence,Southwest University,Chongqing 400715,China
国内会议
济南
英文
1-9
2014-10-16（万方平台首次上网日期，不代表论文的发表时间）