State estimation of nonlinear stochastic systems by evolution strategies based Gaussian sum particle filter

Recently, particle filters have drawn much attention for optimal filtering of nonlinear systems. Particle filters evaluate the grid sum approximation of a posterior probability distribution of the state variable based on observations in Monte Carlo simulation using so-called importance sampling. However, degeneracy phenomena in the importance weights deteriorate the filter performance, and resampling process is introduced to overcome this difficulty. In this paper, we propose a novel Evolution strategies based Gausssian sum filter (ESGSP). It combines the ideas of Gaussian sum filter based on the Gaussian mixture approximation of the posteriori distribution and Evolution strategies based particle filter, in which the selection process in Evolution strategies is substituted into the resampling process in the particle filters. Numerical simulation study indicates the potential to create high performance filters for nonlinear state estimation.