Identification of Time-Varying Systems in Reproducing Kernel Hilbert Spaces

We consider a class of linear systems in state-space form whose parameters evolve in time according to a continuous-time Gauss-Markov process. Our problem is to identify the system from a finite set of noisy output measurements. We derive a connection between this problem and Tikhonov regularization in reproducing kernel Hilbert spaces. Relying upon this result, conditions that ensure the existence of the maximum a posteriori estimate of the parameter trajectory are provided and an identification algorithm is worked out. Simulated data are used to test the goodness of the proposed approach.

[1]  N. Aronszajn.  Theory of Reproducing Kernels. , 1950 .

[2]  Richard Courant,et al.  Methods of Mathematical Physics, 1 , 1955 .

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

[4]  L. Schwartz.  Radon measures on arbitrary topological spaces and cylindrical measures , 1973 .

[5]  A. N. Tikhonov,et al.  Solutions of ill-posed problems , 1977 .

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

[7]  David G. Luenberger,et al.  Linear and nonlinear programming , 1984 .

[8]  G. Wahba.  Spline models for observational data , 1990 .

[9]  Lei Guo.  Estimating time-varying parameters by the Kalman filter based algorithm: stability and convergence , 1990 .

[10]  Costas N. Georghiades,et al.  The expectation-maximization algorithm for symbol unsynchronized sequence detection , 1991, IEEE Trans. Commun..

[11]  A. Dembo,et al.  Onsager-Machlup functionals and maximum a posteriori estimation for a class of non-Gaussian random fields , 1991 .

[12]  Y. Tsypkin,et al.  An optimal algorithm for identification of rapidly time-varying systems , 1992 .

[13]  S. Bittanti,et al.  Bounded error identification of time-varying parameters by RLS techniques , 1994, IEEE Trans. Autom. Control..

[14]  A. Packard.  Gain scheduling via linear fractional transformations , 1994 .

[15]  Marco C. Campi,et al.  Exponentially weighted least squares identification of time-varying systems with white disturbances , 1994, IEEE Trans. Signal Process..

[16]  R. Kohn,et al.  On Gibbs sampling for state space models , 1994 .

[17]  P. Gahinet,et al.  A convex characterization of gain-scheduled H∞ controllers , 1995, IEEE Trans. Autom. Control..

[18]  R. Ravikanth,et al.  Identification of linear parametrically varying systems , 1995, Proceedings of 1995 34th IEEE Conference on Decision and Control.

[19]  Sylvia Richardson,et al.  Markov Chain Monte Carlo in Practice , 1997 .

[20]  L. Wang.  Persistent identification of time-varying systems , 1997, IEEE Trans. Autom. Control..

[21]  Narasimhan Sundararajan,et al.  Identification of time-varying nonlinear systems using minimal radial basis function neural networks , 1997 .

[22]  Giuseppe De Nicolao,et al.  Nonparametric input estimation in physiological systems: Problems, methods, and case studies , 1997, Autom..

[23]  Luo Yingwei,et al.  Identification of time-varying nonlinear systems using minimal radial basis function neural networks : Special section on marine control , 1997 .

[24]  Chung-ki Min.  A Gibbs sampling approach to estimation and prediction of time-varying-parameter models , 1998 .

[25]  Tomaso Poggio,et al.  A Unified Framework for Regularization Networks and Support Vector Machines , 1999 .

[26]  Sean P. Meyn,et al.  Bounds on achievable performance in the identification and adaptive control of time-varying systems , 1999, IEEE Trans. Autom. Control..

[27]  Jamie S. Evans,et al.  Optimal filtering of doubly stochastic auto-regressive processes , 1999, Autom..

[28]  Vikram Krishnamurthy,et al.  Finite dimensional smoothers for MAP state estimation of bilinear systems , 1999, IEEE Trans. Signal Process..

[29]  Arnaud Doucet,et al.  Stochastic sampling algorithms for state estimation of jump Markov linear systems , 2000, IEEE Trans. Autom. Control..

[30]  Christophe Andrieu,et al.  Iterative algorithms for state estimation of jump Markov linear systems , 2001, IEEE Trans. Signal Process..

[31]  H. V. Trees.  Detection, Estimation, And Modulation Theory , 2001 .

[32]  Felipe Cucker,et al.  On the mathematical foundations of learning , 2001 .

[33]  B. Ghosh,et al.  Boundary Location Using Control Theoretic Splines , 2002 .

[34]  Christophe Andrieu,et al.  Efficient particle filtering for Jump Markov Systems , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[35]  Bassam Bamieh,et al.  Identification of linear parameter varying models , 2002 .

[36]  Christophe Andrieu,et al.  Efficient particle filtering for jump Markov systems. Application to time-varying autoregressions , 2003, IEEE Trans. Signal Process..

[37]  Giuseppe De Nicolao,et al.  Regularization networks for inverse problems: A state-space approach , 2003, Autom..

[38]  G. Pillonetto,et al.  Estimating parameters and stochastic functions of one variable using nonlinear measurement models , 2004 .

[39]  Eugenio Cinquemani,et al.  State estimation in stochastic hybrid Systems with sparse observations , 2006, IEEE Transactions on Automatic Control.

[40]  Giuseppe De Nicolao,et al.  Nonparametric identification of population models via Gaussian processes , 2007, Autom..

[41]  Gianluigi Pillonetto,et al.  Solutions of nonlinear control and estimation problems in reproducing kernel Hilbert spaces: Existence and numerical determination , 2008, Autom..