On Stability of a Class of Filters for Nonlinear Stochastic Systems

This article considers stability properties of a broad class of commonly used filters, including the extended and unscented Kalman filters, for discrete and continuous-time stochastic dynamic systems with non-linear state dynamics and linear measurements. We show that contractivity of the filtering error process and boundedness of the error covariance matrix induce stability of the filter. The results are in the form of time-uniform mean square bounds and exponential concentration inequalities for the filtering error. As is typical of stability analysis of filters for non-linear systems, the assumptions required are stringent. However, in contrast to much of the previous analysis, we provide a number of example of model classes for which these assumptions can be verified a priori. Typical requirements are a contractive drift or sufficient inflation of the error covariance matrix and fully observed state. Numerical experiments using synthetic data are used to validate the derived error bounds.

[1]  Wilhelm Stannat,et al.  Long-Time Stability and Accuracy of the Ensemble Kalman-Bucy Filter for Fully Observed Processes and Small Measurement Noise , 2016, SIAM J. Appl. Dyn. Syst..

[2]  Knut Rapp,et al.  Nonlinear estimation and control in the iron ore pelletizing process: An application and analysis of the Extended Kalman Filter , 2004 .

[3]  Ondřej Straka,et al.  Gaussian Process Quadrature Moment Transform , 2017, IEEE Transactions on Automatic Control.

[4]  J. Deyst,et al.  Conditions for asymptotic stability of the discrete minimum-variance linear estimator , 1968 .

[5]  Yuanxin Wu,et al.  A Numerical-Integration Perspective on Gaussian Filters , 2006, IEEE Transactions on Signal Processing.

[6]  Richard S. Bucy,et al.  Global Theory of the Riccati Equation , 1967, J. Comput. Syst. Sci..

[7]  Lars Imsland,et al.  On convergence of the unscented Kalman–Bucy filter using contraction theory , 2016, Int. J. Syst. Sci..

[8]  Pierre Del Moral,et al.  On the Stability of Kalman-Bucy Diffusion Processes , 2016, SIAM J. Control. Optim..

[9]  Simo Särkkä,et al.  On Unscented Kalman Filtering for State Estimation of Continuous-Time Nonlinear Systems , 2007, IEEE Trans. Autom. Control..

[10]  Patric Jensfelt,et al.  A stochastically stable solution to the problem of robocentric mapping , 2009, 2009 IEEE International Conference on Robotics and Automation.

[11]  W. Wonham.  On a Matrix Riccati Equation of Stochastic Control , 1968 .

[12]  Konrad Reif,et al.  Stochastic Stability of the Extended Kalman Filter With Intermittent Observations , 2010, IEEE Transactions on Automatic Control.

[13]  Bernard Delyon,et al.  A note on uniform observability , 2001, IEEE Trans. Autom. Control..

[14]  Catherine Bandle,et al.  Comparison theorems for stochastic differential inequalities and an application to reaction–diffusion equations with random sources , 1994 .

[15]  Konrad Reif,et al.  An EKF-Based Nonlinear Observer with a Prescribed Degree of Stability , 1998, Autom..

[16]  Huang Zhiyuan,et al.  A COMPARISON THEOREM FOR SOLUTIONS OF STOCHASTIC DIFFERENTIAL EQUATIONS AND ITS APPLICATIONS , 1984 .

[17]  B. Anderson,et al.  Detectability and Stabilizability of Time-Varying Discrete-Time Linear Systems , 1981 .

[18]  Rolf Unbehauen,et al.  Modification of the extended Kalman filter with an additive term of instability , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

[19]  R. Unbehauen,et al.  Stochastic stability of the continuous-time extended Kalman filter , 2000 .

[20]  Jean-Jacques E. Slotine,et al.  A Contraction Theory-Based Analysis of the Stability of the Deterministic Extended Kalman Filter , 2015, IEEE Transactions on Automatic Control.

[21]  M. Boutayeb,et al.  Convergence analysis of the extended Kalman filter used as an observer for nonlinear deterministic discrete-time systems , 1997, IEEE Trans. Autom. Control..

[22]  Hugh F. Durrant-Whyte,et al.  A new method for the nonlinear transformation of means and covariances in filters and estimators , 2000, IEEE Trans. Autom. Control..

[23]  Hanxin Zhang,et al.  Modified unscented Kalman filtering and its application in autonomous satellite navigation , 2009 .

[24]  Joseph J. Winkin,et al.  Convergence of the Time-Invariant Riccati Differential Equation towards Its Strong Solution for Stabilizable Systems , 1995 .

[25]  Silverio Bolognani,et al.  Extended Kalman filter tuning in sensorless PMSM drives , 2003 .

[26]  Simo Särkkä,et al.  Fourier-Hermite series for stochastic stability analysis of non-linear Kalman filters , 2016, 2016 19th International Conference on Information Fusion (FUSION).

[27]  Philipp Hennig,et al.  Convergence rates of Gaussian ODE filters , 2018, Statistics and Computing.

[28]  Simo Särkkä,et al.  Gaussian filtering and smoothing for continuous-discrete dynamic systems , 2013, Signal Process..

[29]  Kazufumi Ito,et al.  Gaussian filters for nonlinear filtering problems , 2000, IEEE Trans. Autom. Control..

[30]  Murat Efe,et al.  Stability of the Extended Kalman Filter When the States are Constrained , 2008, IEEE Transactions on Automatic Control.

[31]  Pierre Del Moral,et al.  On the stability and the exponential concentration of Extended Kalman-Bucy filters , 2016 .

[32]  Pierre Del Moral,et al.  On the stability and the uniform propagation of chaos properties of Ensemble Kalman-Bucy filters , 2016, 1605.09329.

[33]  Yuanqing Xia,et al.  Stochastic stability of the unscented Kalman filter with intermittent observations , 2012, Autom..

[34]  T. Ström.  On Logarithmic Norms , 1975 .

[35]  G. Söderlind,et al.  The logarithmic norm. History and modern theory , 2006 .

[36]  Konrad Reif,et al.  Stochastic stability of the discrete-time extended Kalman filter , 1999, IEEE Trans. Autom. Control..

[37]  A. Stuart,et al.  Well-posedness and accuracy of the ensemble Kalman filter in discrete and continuous time , 2013, 1310.3167.

[38]  P. Krishnaprasad,et al.  Dynamic observers as asymptotic limits of recursive filters , 1982, 1982 21st IEEE Conference on Decision and Control.

[39]  Axel Barrau,et al.  The Invariant Extended Kalman Filter as a Stable Observer , 2014, IEEE Transactions on Automatic Control.

[40]  Michel Gevers,et al.  Difference and differential Riccati equations: a note on the convergence to the strong solution , 1992 .

[41]  D. Bernstein.  Matrix Mathematics: Theory, Facts, and Formulas , 2009 .

[42]  Pierre Del Moral,et al.  Mean Field Simulation for Monte Carlo Integration , 2013 .

[43]  A. Krener.  The Convergence of the Extended Kalman Filter , 2002, math/0212255.

[44]  Jouni Hartikainen,et al.  On the relation between Gaussian process quadratures and sigma-point methods , 2015, 1504.05994.

[45]  Pierre Del Moral,et al.  On the Stability and the Uniform Propagation of Chaos of a Class of Extended Ensemble Kalman-Bucy Filters , 2016, SIAM J. Control. Optim..

[46]  Florian Nadel,et al.  Stochastic Processes And Filtering Theory , 2016 .

[47]  C.W. Chan,et al.  Detection of satellite attitude sensor faults using the UKF , 2007, IEEE Transactions on Aerospace and Electronic Systems.

[48]  Gábor Lugosi,et al.  Concentration Inequalities - A Nonasymptotic Theory of Independence , 2013, Concentration Inequalities.

[49]  Yuanxin Wu,et al.  Comments on "Performance evaluation of UKF-based nonlinear filtering" , 2007, Autom..

[50]  Konrad Reif,et al.  The extended Kalman filter as an exponential observer for nonlinear systems , 1999, IEEE Trans. Signal Process..

[51]  D. Talay.  Numerical solution of stochastic differential equations , 1994 .

[52]  Xin T. Tong,et al.  Nonlinear stability and ergodicity of ensemble based Kalman filters , 2015, 1507.08307.

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