Analysis of nonlinear processing ability of sequential Monte Carlo filter

In dealing with the state estimation of nonlinear dynamic systems, the approach usually used is the extented Kalman filter. However, with increasing nonlinear degree of systems, the approach will have greater error rates and dispersing phenomenon. The sequential Monte Carlo filter based on recursive Bayes can estimate the posteriori state of the system by using sample statistics, which can be applied to any nonlinear system, and has strong nonlinear estimation ability. On the basis of discussing the theoretical basis and characteristics of sequential Monte Carlo filtering, this article describes in detail the implementation steps of the algorithm and analyzes the nonlinear processing ability of this method by the simulation of posteriori estimation for nonlinear system.