Application of an improved particle filter for state estimation

Author

Li, Xiang ; Yu, Liu ; Baoku, Su

Author_Institution

Space Control&Inertial Technol. Res. Center, Harbin Inst. of Technol., Harbin

fYear

2008

fDate

16-18 July 2008

Firstpage

489

Lastpage

493

Abstract

A novel Gaussian mixture sigma-point particle filter algorithm is proposed to mitigate the sample depletion problem. The posterior state density is represented by a Gaussian mixture model that is recovered from the weighted particle set of the measurement update step by means of a weighted expectation-maximization algorithm. The simulation results demonstrate the validity of the proposed algorithm.

Keywords

Gaussian processes; Monte Carlo methods; expectation-maximisation algorithm; particle filtering (numerical methods); state estimation; Gaussian mixture model; Gaussian mixture sigma-point particle filter algorithm; measurement update; posterior state density; sample depletion problem; state estimation; weighted expectation-maximization algorithm; weighted particle set; Density measurement; Filtering algorithms; Monte Carlo methods; Nonlinear dynamical systems; Nonlinear equations; Particle filters; Particle measurements; Space technology; State estimation; Vehicle dynamics; Bearing only tracking; Estimation algorithm; Monte carlo simulation; Particle filter;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference, 2008. CCC 2008. 27th Chinese

Conference_Location

Kunming

Print_ISBN

978-7-900719-70-6

Electronic_ISBN

978-7-900719-70-6

Type

conf

DOI

10.1109/CHICC.2008.4604962

Filename

4604962