Particle Flow Particle Filter for Gaussian Mixture Noise Models

Particle filters has become a standard tool for state estimation in nonlinear systems. However, their performance usually deteriorates if the dimension of state space is high or the measurements are highly informative. A major challenge is to construct a proposal density that is well matched to the posterior distribution. Particle flow methods are a promising option for addressing this task. In this paper, we develop a particle flow particle filter algorithm to address the case where both the process noise and the measurement noise are distributed as mixtures of Gaussians. Numerical experiments are performed to explore when the proposed method offers advantages compared to existing techniques.

[1]  A. Doucet,et al.  Gibbs flow for approximate transport with applications to Bayesian computation , 2015, Journal of the Royal Statistical Society: Series B (Statistical Methodology).

[2]  Fred Daum,et al.  Stochastic Particle Flow for Nonlinear High-Dimensional Filtering Problems , 2015, 1511.01448.

[3]  Pete Bunch,et al.  Approximations of the Optimal Importance Density Using Gaussian Particle Flow Importance Sampling , 2014, 1406.3183.

[4]  A. Beskos,et al.  A stable particle filter for a class of high-dimensional state-space models , 2017, Advances in Applied Probability.

[5]  R. Handel,et al.  Can local particle filters beat the curse of dimensionality , 2013, 1301.6585.

[6]  A. Beskos,et al.  On the stability of sequential Monte Carlo methods in high dimensions , 2011, 1103.3965.

[7]  Uwe D. Hanebeck,et al.  Progressive Bayes: a new framework for nonlinear state estimation , 2003, SPIE Defense + Commercial Sensing.

[8]  Mónica F. Bugallo,et al.  Particle filtering for high-dimensional systems , 2013, 2013 5th IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP).

[9]  Fred Daum,et al.  Exact particle flow for nonlinear filters , 2010, Defense + Commercial Sensing.

[10]  Gareth W. Peters,et al.  Langevin and Hamiltonian Based Sequential MCMC for Efficient Bayesian Filtering in High-Dimensional Spaces , 2015, IEEE Journal of Selected Topics in Signal Processing.

[11]  N. Gordon,et al.  Novel approach to nonlinear/non-Gaussian Bayesian state estimation , 1993 .

[12]  Sebastian Reich,et al.  A Guided Sequential Monte Carlo Method for the Assimilation of Data into Stochastic Dynamical Systems , 2013 .

[13]  Mónica F. Bugallo,et al.  Multiple Particle Filtering , 2007, 2007 IEEE International Conference on Acoustics, Speech and Signal Processing - ICASSP '07.

[14]  Martin Ulmke,et al.  Non-linear and non-Gaussian state estimation using log-homotopy based particle flow filters , 2014, 2014 Sensor Data Fusion: Trends, Solutions, Applications (SDF).

[15]  Wolfgang Koch,et al.  A log homotopy based particle flow solution for mixture of Gaussian prior densities , 2016, 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI).

[16]  Mark Coates,et al.  Gaussian sum particle flow filter , 2017, 2017 IEEE 7th International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP).

[17]  Petar M. Djuric,et al.  Gaussian sum particle filtering , 2003, IEEE Trans. Signal Process..

[18]  Yunpeng Li,et al.  Particle Filtering With Invertible Particle Flow , 2016, IEEE Transactions on Signal Processing.

[19]  Tao Ding,et al.  Implementation of the Daum-Huang exact-flow particle filter , 2012, 2012 IEEE Statistical Signal Processing Workshop (SSP).