Resampling schemes for Rao-Blackwellization Particle Filters

Particle filter is a general numerical algorithm to estimate the posterior probability density function. While the execution of particle filter is very simple, but its principle drawback is that, it is very computational expensive by increasing rapidly with the increasing in the state measurement. One solution for this issue is to minimize the states showing up directly in the flow. The primary commitment in this paper is to determine the subtle elements for the Rao-Blackwellization (marginalize) Particle Filter (RBPF) for a general nonlinear and non-gaussian state-space model. In this article a comparison is made between commonly encountered resampling algorithms for particle filters and Rao-Blackwellization (marginalize) Particle Filter (RBPF). This facilitates a similarity of the algorithms with respect to their resampling quality. Using wide-ranging Monte Carlo simulations the theoretical outcomes are verified.