A rank correlation coefficient based particle filter to estimate parameters in non-linear models

Qingxu Meng,Kaicheng Li

doi:10.1177/1550147719841273

Abstract

Particle filtering algorithm has found an increasingly wide utilization in many fields at present, especially in non-linear and non-Gaussian situations. Because of the particle degeneracy limitatio...

Highlights

Since Gordon et al.[1] research in 1993, particle filtering (PF) algorithm known as sequential Monte Carlo (SMC) method has become a recent technique to perform filtering and smoothing for non-linear and nonGaussian systems
A new particle filter algorithm combing with different rank correlation coefficients is proposed
The computational complexity of the proposed algorithm is lower than Gaussian sum particle filter (GSPF) and Gaussian mixture sigma-point particle filter (GMSPPF), which can be reflected by time consumption in Gaussian mixture noise

Summary

Introduction

Since Gordon et al.[1] research in 1993, particle filtering (PF) algorithm known as sequential Monte Carlo (SMC) method has become a recent technique to perform filtering and smoothing for non-linear and nonGaussian systems. The key idea of this algorithm is to construct the posterior density function (pdf) of the state variables by a set of random samples (particles) with associated weights recursively. The idea of traditional PF algorithm is to draw samples (particles) from the PDF p(xkjy1:k) theoretically and set the weights to be equal.

Results

Conclusion