A linear programming approach to online set membership parameter estimation for linear regression models

This paper presents a new technique for online set membership parameter estimation of linear regression models affected by unknown-but-bounded noise. An orthotopic approximation of the set of feasible parameters is updated at each time step. The proposed technique relies on the solution of a suitable linear program, whenever a new measurement leads to a reduction of the approximating orthotope. The key idea for preventing the size of the linear programs from steadily increasing is to propagate only the binding constraints of these optimization problems. Numerical studies show that the new approach outperforms existing recursive set approximation techniques, while keeping the required computational burden within the same order of magnitude. Copyright © 2016 John Wiley & Sons, Ltd.

[1]  Dario Piga,et al.  Computational Load Reduction in Bounded Error Identification of Hammerstein Systems , 2013, IEEE Transactions on Automatic Control.

[2]  Gustavo Belforte,et al.  Parameter estimation algorithms for a set-membership description of uncertainty , 1990, Autom..

[3]  Rolf Isermann,et al.  Fault-diagnosis systems : an introduction from fault detection to fault tolerance , 2006 .

[4]  J. R. Deller,et al.  Unifying the landmark developments in optimal bounding ellipsoid identification , 1994 .

[5]  J. Norton,et al.  Bounding Approaches to System Identification , 1996 .

[6]  Dario Piga,et al.  A convex relaxation approach to set-membership identification of LPV systems , 2012, Autom..

[7]  Gilles Ferreres,et al.  Estimation of output error models in the presence of unknown but bounded disturbances , 1997 .

[8]  Y. F. Huang,et al.  On the value of information in system identification - Bounded noise case , 1982, Autom..

[9]  A. Kurzhanski.  Identification - a theory of guaranteed estimates , 1989 .

[10]  E. Walter,et al.  Estimation of parameter bounds from bounded-error data: a survey , 1990 .

[11]  A. Vicino,et al.  Sequential approximation of feasible parameter sets for identification with set membership uncertainty , 1996, IEEE Trans. Autom. Control..

[12]  Andrea Garulli,et al.  A constraint selection technique for recursive set membership identification , 2014 .

[13]  Andrea Garulli,et al.  On worst-case approximation of feasible system sets via orthonormal basis functions , 2003, IEEE Trans. Autom. Control..

[14]  Fred C. Schweppe,et al.  Uncertain dynamic systems , 1973 .

[15]  V. Cerone,et al.  Bounded error identification of Hammerstein systems through sparse polynomial optimization , 2012, Autom..

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

[17]  A. Garulli,et al.  Block recursive parallelotopic bounding in set membership identification , 1998 .

[18]  Vicenç Puig,et al.  Robust fault detection using polytope-based set-membership consistency test , 2012 .

[19]  Jean B. Lasserre,et al.  A Unified Framework for Solving a General Class of Conditional and Robust Set-Membership Estimation Problems , 2014, IEEE Transactions on Automatic Control.

[20]  D. Bertsekas,et al.  Recursive state estimation for a set-membership description of uncertainty , 1971 .

[21]  Constantino M. Lagoa,et al.  Convex Certificates for Model (In)validation of Switched Affine Systems With Unknown Switches , 2014, IEEE Transactions on Automatic Control.

[22]  V. Broman,et al.  A compact algorithm for the intersection and approximation of N -dimensional polytopes , 1990 .

[23]  Karl Johan Åström,et al.  Adaptive Control , 1989, Embedded Digital Control with Microcontrollers.

[24]  Andrea Garulli,et al.  A constraint selection technique for set membership estimation of time-varying parameters , 2014, 53rd IEEE Conference on Decision and Control.

[25]  E. K. Kostousova.  State estimation for dynamic systems via parallelotopes optimization and parallel computations , 1998 .

[26]  Vicenç Puig,et al.  Identification for passive robust fault detection using zonotope‐based set‐membership approaches , 2011 .

[27]  Eduardo F. Camacho,et al.  Bounded error identification of systems with time-varying parameters , 2004, 2004 43rd IEEE Conference on Decision and Control (CDC) (IEEE Cat. No.04CH37601).

[28]  Narendra Karmarkar,et al.  A new polynomial-time algorithm for linear programming , 1984, Comb..

[29]  Constantino M. Lagoa,et al.  Set membership identification of switched linear systems with known number of subsystems , 2015, Autom..

[30]  Dario Piga,et al.  Improved parameter bounds for set‐membership EIV problems , 2011 .

[31]  Gustavo Belforte,et al.  Two new estimation algorithms for linear models with unknown but bounded measurement noise , 1993, IEEE Trans. Autom. Control..

[32]  Andrea Garulli,et al.  Input design in worst-case system identification with quantized measurements , 2012, Autom..

[33]  Constantino M. Lagoa,et al.  A convex optimization approach to model (in)validation of switched ARX systems with unknown switches , 2012, 2012 IEEE 51st IEEE Conference on Decision and Control (CDC).

[34]  Boris T. Polyak,et al.  Special issue on the set membership modelling of uncertainties in dynamical systems , 2005 .

[35]  Antonio Vicino,et al.  Estimation theory for nonlinear models and set membership uncertainty , 1991, Autom..

[36]  Michael Nikolaou,et al.  Simultaneous Constrained Model Predictive Control and Identification of DARX Processes , 1998, Autom..

[37]  Andrea Garulli,et al.  Feasible Parameter Set Approximation for Linear Models with Bounded Uncertain Regressors , 2014, IEEE Transactions on Automatic Control.