Quantitative verification of Kalman filters

Kalman filters are widely used for estimating the state of a system based on noisy or inaccurate sensor readings, for example in the control and navigation of vehicles or robots. However, numerical instability or modelling errors may lead to divergence of the filter, leading to erroneous estimations. Establishing robustness against such issues can be challenging. We propose novel formal verification techniques and software to perform a rigorous quantitative analysis of the effectiveness of Kalman filters. We present a general framework for modelling Kalman filter implementations operating on linear discrete-time stochastic systems, and techniques to systematically construct a Markov model of the filter’s operation using truncation and discretisation of the stochastic noise model. Numerical stability and divergence properties are then verified using probabilistic model checking. We evaluate the scalability and accuracy of our approach on two distinct probabilistic kinematic models and four Kalman filter implementations.

[1]  Chenglin Wen,et al.  Performance Analysis of the Kalman Filter With Mismatched Noise Covariances , 2016, IEEE Transactions on Automatic Control.

[2]  Grigore Rosu,et al.  Certifying Optimality of State Estimation Programs , 2003, CAV.

[3]  Luiz Paulo Lopes Fávero,et al.  Estimation , 2019, Data Science for Business and Decision Making.

[4]  R. E. Kalman,et al.  A New Approach to Linear Filtering and Prediction Problems , 2002 .

[5]  D. Muijs,et al.  4th Edition , 2006 .

[6]  Bengt Jonsson,et al.  A logic for reasoning about time and reliability , 1990, Formal Aspects of Computing.

[7]  N. A. Carlson.  Fast triangular formulation of the square root filter. , 1973 .

[8]  G. Trenkler.  Continuous univariate distributions , 1994 .

[9]  Y. Bar-Shalom.  Tracking and data association , 1988 .

[10]  Alexandros Evangelidis,et al.  Quantitative Verification of Numerical Stability for Kalman Filters , 2019, FM.

[11]  Renato Zanetti,et al.  Information Formulation of the UDU Kalman Filter , 2019, IEEE Transactions on Aerospace and Electronic Systems.

[12]  D. Hermes,et al.  Navigation , 2019, Building Xamarin.Forms Mobile Apps Using XAML.

[13]  Thiagalingam Kirubarajan,et al.  Estimation with Applications to Tracking and Navigation , 2001 .

[14]  Larry Wasserman,et al.  All of Statistics: A Concise Course in Statistical Inference , 2004 .

[15]  Robert Grover Brown,et al.  Introduction to random signals and applied Kalman filtering : with MATLAB exercises and solutions , 1996 .

[16]  A. Bryson,et al.  Discrete square root filtering: A survey of current techniques , 1971 .

[17]  Thomas Kailath,et al.  Linear Systems , 1980 .

[18]  Marta Z. Kwiatkowska,et al.  Stochastic Model Checking , 2007, SFM.

[19]  A.H. Haddad,et al.  Applied optimal estimation , 1976, Proceedings of the IEEE.

[20]  G. Bierman.  Measurement updating using the U-D factorization , 1975 .

[21]  Peter S. Maybeck,et al.  Stochastic Models, Estimation And Control , 2012 .

[22]  D. Simon.  Kalman filtering with state constraints: a survey of linear and nonlinear algorithms , 2010 .

[23]  Paul Zarchan,et al.  Fundamentals of Kalman Filtering: A Practical Approach , 2001 .

[24]  Greg Welch,et al.  An Introduction to Kalman Filter , 1995, SIGGRAPH 2001.

[25]  M. Kwiatkowska.  Quantitative verification: models, techniques and tools , 2007, ESEC-FSE companion '07.

[26]  Paul Zarchan,et al.  Fundamentals of Kalman Filtering: A Practical Approach, Fourth Edition , 2015 .

[27]  Ian R. Reid.  Estimation II , 2010 .

[28]  H. Nicholson,et al.  Control of divergence in Kalman filters , 1976 .

[29]  Bruce P. Gibbs,et al.  Advanced Kalman Filtering, Least-Squares and Modeling: A Practical Handbook , 2011 .

[30]  G. Bierman.  Factorization methods for discrete sequential estimation , 1977 .

[31]  On the Verification of Synthesized Kalman Filters , 2003 .

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

[33]  LI X.RONG,et al.  Survey of maneuvering target tracking. Part I. Dynamic models , 2003 .

[34]  S. Chiu,et al.  On-Line Detection & Correction of Kalman Filter Divergence by Fuzzy Logic , 1993, 1993 American Control Conference.

[35]  X. R. Li,et al.  Survey of maneuvering target tracking. Part I. Dynamic models , 2003 .

[36]  Johann Schumann,et al.  Automating the implementation of Kalman filter algorithms , 2004, TOMS.

[37]  T. Schlick,et al.  Supporting Material , 2006 .

[38]  Marta Z. Kwiatkowska,et al.  PRISM 4.0: Verification of Probabilistic Real-Time Systems , 2011, CAV.

[39]  J. A. Salvato.  John wiley & sons. , 1994, Environmental science & technology.

[40]  Charles M. Grinstead,et al.  Introduction to probability , 1999, Statistics for the Behavioural Sciences.

[41]  Mark Moulin,et al.  Formal Verification of Maneuvering Target Tracking , 2003 .

[42]  A. Jazwinski.  Stochastic Processes and Filtering Theory , 1970 .

[43]  Marta Z. Kwiatkowska,et al.  Automated Verification Techniques for Probabilistic Systems , 2011, SFM.

[44]  Petros G. Voulgaris,et al.  On optimal ℓ∞ to ℓ∞ filtering , 1995, Autom..

[45]  P. Dooren,et al.  Numerical aspects of different Kalman filter implementations , 1986 .

[46]  Mohinder S. Grewal,et al.  Kalman Filtering: Theory and Practice Using MATLAB , 2001 .

[47]  Thia Kirubarajan,et al.  Estimation with Applications to Tracking and Navigation: Theory, Algorithms and Software , 2001 .