A NEW SUBSPACE IDENTIFICATION METHOD FOR OPEN AND CLOSED LOOP DATA

Abstract Subspace methods have emerged as useful tools for the identification of linear time invariant discrete time systems. Most of the methods have been developed for the open loop case to avoid difficulties with data correlations due to the feedback. This paper extends some recent ideas for developing subspace methods that can perform well on data collected both in open and closed loop conditions. Here, a method that aims at minimizing the prediction errors in several approximate steps is proposed. The steps involve using constrained least squares estimation on models with different degrees of structure such as block-toeplitz, and reduced rank matrices. The statistical estimation performance of the method is shown to be competitive to existing subspace methods in a simulation example.

[1]  Lennart Ljung,et al.  Closed-Loop Identification Revisited - Updated Version , 1998 .

[2]  Lennart Ljung,et al.  Closed-Loop Subspace Identification with Innovation Estimation , 2003 .

[3]  Michel Verhaegen,et al.  Application of a subspace model identification technique to identify LTI systems operating in closed-loop , 1993, Autom..

[4]  Magnus Jansson,et al.  Subspace Identification and ARX Modeling , 2003 .

[5]  Alessandro Chiuso,et al.  Consistency analysis of closed-loop subspace identification , 2004, 2004 43rd IEEE Conference on Decision and Control (CDC) (IEEE Cat. No.04CH37601).

[6]  Tony Gustafsson.  Subspace identification using instrumental variable techniques , 2001, Autom..

[7]  Alessandro Chiuso.  Asymptotic variance of a certain closed-loop subspace identification method , 2004, 2004 43rd IEEE Conference on Decision and Control (CDC) (IEEE Cat. No.04CH37601).

[8]  Manfred Deistler,et al.  Statistical analysis of novel subspace identification methods , 1996, Signal Process..

[9]  B. De Moor,et al.  Closed loop subspace system identification , 1997, Proceedings of the 36th IEEE Conference on Decision and Control.

[10]  Michel Verhaegen,et al.  Closed-loop identification using canonical correlation analysis , 1999, 1999 European Control Conference (ECC).

[11]  Magnus Jansson,et al.  Weighted low rank approximation and reduced rank linear regression , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[12]  Lennart Ljung,et al.  Closed-loop identification revisited , 1999, Autom..

[13]  Dietmar Bauer,et al.  Order estimation for subspace methods , 2001, Autom..

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

[15]  Lennart Ljung,et al.  Subspace identification from closed loop data , 1996, Signal Process..

[16]  W. Larimore.  System Identification, Reduced-Order Filtering and Modeling via Canonical Variate Analysis , 1983, 1983 American Control Conference.

[17]  Michel Verhaegen,et al.  Identification of the deterministic part of MIMO state space models given in innovations form from input-output data , 1994, Autom..