State Estimation for General Class of Dynamical Systems: An Extension to Particle Filters

Many physical systems are nonlinear and non-Gaussian in their state-space models. Particle Filter (PF) is a sequential Monte Carlo method that uses sets of sample scenarios, i.e. “particles” to represent probability densities, and it can be applied for state estimation in nonlinear/non-Gaussian state-spaces models. Conventional variants of PF do not assume any noise for the system input, while the corresponding measurement models disregard the system input as an argument. In reality, physical systems receive inputs contaminated with the measurement noise. In this work, a generalized particle filter algorithm is developed that handles the noisy input of the state-space model in a probabilistic framework. Three advanced variants of PF are then developed to improve the filtering accuracy. Performance of the developed filters are then verified with simulation of univariate and bivariate non-stationary growth models as benchmarks.

[1]  Jie Liu,et al.  A regularized auxiliary particle filtering approach for system state estimation and battery life prediction , 2011 .

[2]  F Gustafsson,et al.  Particle filter theory and practice with positioning applications , 2010, IEEE Aerospace and Electronic Systems Magazine.

[3]  Khashayar Khorasani,et al.  A novel particle filter parameter prediction scheme for failure prognosis , 2014, 2014 American Control Conference.

[4]  Kristine L. Bell,et al.  A Tutorial on Particle Filters for Online Nonlinear/NonGaussian Bayesian Tracking , 2007 .

[5]  Ying Chen,et al.  Sequential state estimation of nonlinear/non-Gaussian systems with stochastic input for turbine degradation estimation , 2016 .

[6]  T. Başar,et al.  A New Approach to Linear Filtering and Prediction Problems , 2001 .

[7]  P. Fearnhead,et al.  Improved particle filter for nonlinear problems , 1999 .

[8]  S. Ramamurthy,et al.  Digital Filters and Signal Processing with MATLAB Exercises , 2014 .

[9]  M. Pitt,et al.  Filtering via Simulation: Auxiliary Particle Filters , 1999 .

[10]  Ying Chen,et al.  A framework with nonlinear system model and nonparametric noise for gas turbine degradation state estimation , 2015 .

[11]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

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

[13]  P. V. Kerm,et al.  Adaptive kernel density estimation , 2003 .

[14]  D. Simon.  Optimal State Estimation: Kalman, H Infinity, and Nonlinear Approaches , 2006 .

[15]  Nando de Freitas,et al.  Sequential Monte Carlo Methods in Practice , 2001, Statistics for Engineering and Information Science.

[16]  Farid Golnaraghi,et al.  A Kalman/Particle Filter-Based Position and Orientation Estimation Method Using a Position Sensor/Inertial Measurement Unit Hybrid System , 2010, IEEE Transactions on Industrial Electronics.

[17]  Neil J. Gordon,et al.  Editors: Sequential Monte Carlo Methods in Practice , 2001 .

[18]  N. Oudjane,et al.  Progressive correction for regularized particle filters , 2000, Proceedings of the Third International Conference on Information Fusion.

[19]  Walter Sextro,et al.  PEM fuel cell prognostics using particle filter with model parameter adaptation , 2014, 2014 International Conference on Prognostics and Health Management.

[20]  Mohammad Teshnehlab,et al.  A SLAM based on auxiliary marginalised particle filter and differential evolution , 2014, Int. J. Syst. Sci..

[21]  Jeffrey K. Uhlmann,et al.  New extension of the Kalman filter to nonlinear systems , 1997, Defense, Security, and Sensing.