Modeling and identification of uncertain-input systems

In this work, we present a new class of models, called uncertain-input models, that allows us to treat system-identification problems in which a linear system is subject to a partially unknown input signal. To encode prior information about the input or the linear system, we use Gaussian-process models. We estimate the model from data using the empirical Bayes approach: the input and the impulse responses of the linear system are estimated using the posterior means of the Gaussian-process models given the data, and the hyperparameters that characterize the Gaussian-process models are estimated from the marginal likelihood of the data. We propose an iterative algorithm to find the hyperparameters that relies on the EM method and results in simple update steps. In the most general formulation, neither the marginal likelihood nor the posterior distribution of the unknowns is tractable. Therefore, we propose two approximation approaches, one based on Markov-chain Monte Carlo techniques and one based on variational Bayes approximation. We also show special model structures for which the distributions are treatable exactly. Through numerical simulations, we study the application of the uncertain-input model to the identification of Hammerstein systems and cascaded linear systems. As part of the contribution of the paper, we show that this model structure encompasses many classical problems in system identification such as classical PEM, Hammerstein models, errors-in-variables problems, blind system identification, and cascaded linear systems. This allows us to build a systematic procedure to apply the algorithms proposed in this work to a wide class of classical problems.

[1]  Karim Abed-Meraim,et al.  Blind system identification , 1997, Proc. IEEE.

[2]  H. Harry Asada,et al.  Laguerre-model blind system identification: cardiovascular dynamics estimated from multiple peripheral circulatory signals , 2005, IEEE Transactions on Biomedical Engineering.

[3]  Ronald C. Neath,et al.  On Convergence Properties of the Monte Carlo EM Algorithm , 2012, 1206.4768.

[4]  Thomas B. Schön,et al.  A flexible state space model for learning nonlinear dynamical systems , 2016, Autom..

[5]  Henrik Ohlsson,et al.  Blind Identification via Lifting , 2013, ArXiv.

[6]  Stephen A. Billings,et al.  Identi cation of a class of nonlinear systems using correlation analysis , 1978 .

[7]  Torsten Söderström System identification for the errors-in-variables problem , 2010 .

[8]  Giulio Bottegal,et al.  Kernel-based system identification from noisy and incomplete input-output data , 2016, 2016 IEEE 55th Conference on Decision and Control (CDC).

[9]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[10]  Rik Pintelon,et al.  Errors-in-variables identification of dynamic systems in general cases , 2015 .

[11]  Karl Henrik Johansson,et al.  Estimation of building occupancy levels through environmental signals deconvolution , 2013, BuildSys@SenSys.

[12]  Torsten Söderström,et al.  Why are errors-in-variables problems often tricky? , 2003, ECC.

[13]  Giulio Bottegal,et al.  On the estimation of initial conditions in kernel-based system identification , 2015, 2015 54th IEEE Conference on Decision and Control (CDC).

[14]  Jonas Linder,et al.  Identification of systems with unknown inputs using indirect input measurements , 2017, Int. J. Control.

[15]  Giulio Bottegal,et al.  Blind system identification using kernel-based methods , 2014, ArXiv.

[16]  G. C. Wei,et al.  A Monte Carlo Implementation of the EM Algorithm and the Poor Man's Data Augmentation Algorithms , 1990 .

[17]  Riccardo Sven Risuleo System identification with input uncertainties : an EM kernel-based approach , 2016 .

[18]  Stochastic Relaxation , 2014, Computer Vision, A Reference Guide.

[19]  Francesco Dinuzzo,et al.  Kernels for Linear Time Invariant System Identification , 2012, SIAM J. Control. Optim..

[20]  L. Tong,et al.  Multichannel blind identification: from subspace to maximum likelihood methods , 1998, Proc. IEEE.

[21]  Donald Geman,et al.  Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images , 1984 .

[22]  E. Bai An optimal two stage identification algorithm for Hammerstein-Wiener nonlinear systems , 1998 .

[23]  Håkan Hjalmarsson,et al.  System identification of complex and structured systems , 2009, 2009 European Control Conference (ECC).

[24]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[25]  Giulio Bottegal,et al.  A new kernel-based approach to overparameterized Hammerstein system identification , 2015, 2015 54th IEEE Conference on Decision and Control (CDC).

[26]  G. McLachlan,et al.  The EM algorithm and extensions , 1996 .

[27]  L. Ljung,et al.  Constructive state space model induced kernels for regularized system identification , 2014 .

[28]  L. Ljung,et al.  Maximum Likelihood Identification of Wiener Models , 2008 .

[29]  Ivan Markovsky,et al.  Structured Low-Rank Approximation with Missing Data , 2013, SIAM J. Matrix Anal. Appl..

[30]  Anders Hansson,et al.  Maximum likelihood estimation of linear SISO models subject to missing output data and missing input data , 2014, Int. J. Control.

[31]  E. Bai,et al.  Block Oriented Nonlinear System Identification , 2010 .

[32]  Giulio Bottegal,et al.  A kernel-based approach to Hammerstein system identification , 2014, ArXiv.

[33]  W. Greblicki,et al.  Identification of discrete Hammerstein systems using kernel regression estimates , 1986 .

[34]  Er-Wei Bai,et al.  Convergence of the iterative Hammerstein system identification algorithm , 2004, 2004 43rd IEEE Conference on Decision and Control (CDC) (IEEE Cat. No.04CH37601).

[35]  Biao Huang,et al.  System Identification , 2000, Control Theory for Physicists.

[36]  Alessandro Chiuso,et al.  A Bayesian learning approach to linear system identification with missing data , 2009, Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference.

[37]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[38]  Carl E. Rasmussen,et al.  Identification of Gaussian Process State-Space Models with Particle Stochastic Approximation EM , 2013, ArXiv.

[39]  Matthew J. Beal Variational algorithms for approximate Bayesian inference , 2003 .

[40]  Nobuharu Nakajima,et al.  Blind deconvolution using the maximum likelihood estimation and the iterative algorithm , 1993 .

[41]  Henrik Ohlsson,et al.  On the estimation of transfer functions, regularizations and Gaussian processes - Revisited , 2012, Autom..

[42]  Lennart Ljung,et al.  Identification of Hammerstein-Wiener models , 2013, Autom..

[43]  Håkan Hjalmarsson,et al.  A geometric approach to variance analysis of cascaded systems , 2013, 52nd IEEE Conference on Decision and Control.

[44]  Er-Wei Bai,et al.  On the Least Squares Solutions of a System of Bilinear Equations , 2005, Proceedings of the 44th IEEE Conference on Decision and Control.

[45]  itannesven Identification of dynamic errors-in-variables models , 2019 .

[46]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[47]  Peter Green,et al.  Markov chain Monte Carlo in Practice , 1996 .

[48]  T. Söderström ERRORS-IN-VARIABLES METHODS IN SYSTEM IDENTIFICATION , 2006 .

[49]  New York Dover,et al.  ON THE CONVERGENCE PROPERTIES OF THE EM ALGORITHM , 1983 .

[50]  R. A. Leibler,et al.  On Information and Sufficiency , 1951 .

[51]  R. Wolpert,et al.  Integrated likelihood methods for eliminating nuisance parameters , 1999 .

[52]  Xiao-Li Meng,et al.  Maximum likelihood estimation via the ECM algorithm: A general framework , 1993 .

[53]  Lennart Ljung,et al.  Kernel methods in system identification, machine learning and function estimation: A survey , 2014, Autom..

[54]  Justin K. Romberg,et al.  Blind Deconvolution Using Convex Programming , 2012, IEEE Transactions on Information Theory.

[55]  Giuseppe De Nicolao,et al.  A new kernel-based approach for system identification , 2008, 2008 American Control Conference.

[56]  Dennis V. Lindley,et al.  Empirical Bayes Methods , 1974 .

[57]  J. S. Maritz,et al.  Empirical Bayes Methods , 1974 .

[58]  Lennart Ljung,et al.  Developments in The MathWorks System Identification Toolbox , 2009 .

[59]  Radford M. Neal Pattern Recognition and Machine Learning , 2007, Technometrics.

[60]  Lennart Ljung,et al.  Revisiting the Two-Stage Algorithm for Hammerstein system identification , 2009, Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference.

[61]  Giuseppe De Nicolao,et al.  A new kernel-based approach for linear system identification , 2010, Autom..

[62]  Eric Moulines,et al.  Subspace methods for the blind identification of multichannel FIR filters , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.