An improved particle swarm optimization particle filter algorithm based on harmony search

Abstract

In light of the problems that particle filter based on particle swarm optimization (PSO-PF) algorithm is easy to fall into the local optimum in the state estimation of complex nonlinear systems, an improved particle swarm optimization particle filter algorithm based on harmony search (HSPSO-PF) is proposed. The optimal value of each individual in the particle swarm algorithm is equivalent to a variable in the harmony memory (HM) in the harmony search algorithm. On the one hand, the idea of generating solutions in harmony search algorithm is used to update the matrix that is composed of particles' optimal solutions in the iteration of PSO. And the harmony memory consideration rate is adjusted to ensure that the source of solution generation is not limited to the HM, so that the PSO can search for the global optimal solution. On the other hand, the pitch adjustment rate and the fine-tuning bandwidth are adjusted to improve the local search ability of the PSO. Simulation results under Gaussian and non-Gaussian show that HSPSO-PF can effectively improve the estimation accuracy of the particle filtering algorithm, and has extensive adaptability.