Emperical Study on the Effect of Multi-Sampling in the Prediction Step of the Particle Filter

TL;DR

This paper empirically investigates how multi-sampling in the prediction step of particle filters affects the accuracy of smoothed distribution estimates, addressing particle degeneracy issues.

Contribution

It introduces and tests the use of multiple particles in the prediction step of particle filters, providing empirical evidence of its impact on performance.

Findings

Multi-sampling improves the accuracy of smoothed distribution estimates.

Multi-sampling reduces particle degeneracy in particle filters.

Empirical results demonstrate enhanced performance with multi-sampling.

Abstract

Particle filters are applicable to a wide range of nonlinear, non-Gaussian state-space models and have already been applied to a variety of problems. However, there is a problem in the calculation of smoothed distributions, where particles gradually degenerate and accuracy is reduced. The purpose of this paper is to consider the possibility of generating multiple particles in the prediction step of the particle filter and to empirically verify the effect using real data.