Optimal asymptotic identification under bounded disturbances

The intrinsic limitation of worst-case identification of linear time-invariant systems using data corrupted by bounded disturbances, when the unknown plant is known to belong to a given model set, is studied. This is done by analyzing the optimal worst-case asymptotic error achievable by performing experiments using any bounded input and estimating the plant using any identification algorithm. It is shown that under some topological conditions on the model set, there is an identification algorithm which is asymptotically optimal for any input, and the optimal asymptotic error is characterized as a function of the inputs. These results, which hold for any error metric and disturbance norm, are applied to three specific identification problems: identification of stable systems in the l/sub 1/ norm, identification of stable rational systems in the H/sub infinity / norm and identification of unstable rational systems in the gap metric. For each of these problems, the general characterization of optimal asymptotic error is used to find near-optimal inputs to minimize the error. >

[1]  A. El-Sakkary,et al.  The gap metric: Robustness of stabilization of feedback systems , 1985 .

[2]  G. Zames.  On the metric complexity of casual linear systems: ε -Entropy and ε -Dimension for continuous time , 1979 .

[3]  John C. Doyle.  Analysis of Feedback Systems with Structured Uncertainty , 1982 .

[4]  James R. Munkres,et al.  Topology; a first course , 1974 .

[5]  Henryk Wozniakowski,et al.  Information-based complexity , 1987, Nature.

[6]  Arthur J. Helmicki,et al.  Identification in H ∞ : linear algorithms , 1990 .

[7]  C. Jacobson,et al.  Worst Case System Identification in l1: Optimal Algorithms and Error Bounds , 1991, 1991 American Control Conference.

[8]  Mathukumalli Vidyasagar,et al.  Control System Synthesis , 1985 .

[9]  David N. C Tse.  Optimal and robust identification under bounded disturbances , 1991 .

[10]  A. Helmicki,et al.  Identification in H∞: a robustly convergent, nonlinear algorithm , 1990, 1990 American Control Conference.

[11]  J. Tsitsiklis,et al.  The sample complexity of worst-case identification of FIR linear systems , 1993, Proceedings of 32nd IEEE Conference on Decision and Control.

[12]  David Haussler,et al.  Occam's Razor , 1987, Inf. Process. Lett..

[13]  Charles A. Micchelli,et al.  A Survey of Optimal Recovery , 1977 .

[14]  M. Milanese,et al.  Estimation theory and uncertainty intervals evaluation in presence of unknown but bounded errors: Linear families of models and estimators , 1982 .

[15]  J. P. Norton,et al.  Identification and application of bounded-parameter models , 1985, Autom..

[16]  P. Khargonekar,et al.  Linear and nonlinear algorithms for identification in H/sub infinity / with error bounds , 1992 .

[17]  Stephen P. Boyd,et al.  Parameter set estimation of systems with uncertain nonparametric dynamics and disturbances , 1990, 29th IEEE Conference on Decision and Control.

[18]  M.A. Dahleh,et al.  Robust and Optimal Identification in the l1 Norm , 1991, 1991 American Control Conference.

[19]  Leslie G. Valiant,et al.  A theory of the learnable , 1984, STOC '84.

[20]  Judea Pearl,et al.  ON THE CONNECTION BETWEEN THE COMPLEXITY AND CREDIBILITY OF INFERRED MODELS , 1978 .

[21]  Eric Walter,et al.  Experiment design in a bounded-error context: Comparison with D-optimality , 1989, Autom..

[22]  M. Dahleh,et al.  A necessary and sufficient condition for robust BIBO stability , 1988 .

[23]  J. Partington,et al.  Robust approximation and identification in H∞ , 1991, 1991 American Control Conference.

[24]  G. Stein,et al.  Robust performance of adaptive controllers with general uncertainty structure , 1990, 29th IEEE Conference on Decision and Control.

[25]  P. Mäkilä.  Robust identification and Galois sequences , 1991 .

[26]  R. A. Silverman,et al.  Theory of Functions of a Complex Variable , 1968 .

[27]  Carl N. Nett,et al.  Control oriented system identification: a worst-case/deterministic approach in H/sub infinity / , 1991 .

[28]  Cornelis Praagman,et al.  A lower and upper bound for the gap metric , 1989, Proceedings of the 28th IEEE Conference on Decision and Control,.

[29]  Leszek Plaskota,et al.  Noisy information for linear problems in the asymptotic setting , 1991, J. Complex..

[30]  G. Zames,et al.  Uncertainty Principles and Identification n-Widths for LTI and Slowly Varying Systems , 1992, 1992 American Control Conference.

[31]  T. Georgiou,et al.  Optimal robustness in the gap metric , 1990 .

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

[33]  Raman K. Mehra,et al.  Optimal input signals for parameter estimation in dynamic systems--Survey and new results , 1974 .

[34]  G. Zames,et al.  Fast identification n-widths and uncertainty principles for LTI and slowly varying systems , 1994, IEEE Trans. Autom. Control..

[35]  Antonio Vicino,et al.  Strongly optimal algorithms and optimal information in estimation problems , 1986, J. Complex..

[36]  Rogelio Lozano,et al.  Reformulation of the parameter identification problem for systems with bounded disturbances , 1987, Autom..

[37]  R. Tempo,et al.  Optimal algorithms theory for robust estimation and prediction , 1985 .

[38]  Munther A. Dahleh,et al.  Optimal rejection of persistent disturbances, robust stability, and mixed sensitivity minimization , 1988 .

[39]  K. Poolla,et al.  On the time complexity of worst-case system identification , 1994, IEEE Trans. Autom. Control..

[40]  M. Khammash,et al.  Performance robustness of discrete-time systems with structured uncertainty , 1991 .

[41]  Henryk Wozniakowski,et al.  A general theory of optimal algorithms , 1980, ACM monograph series.

[42]  M. Milanese,et al.  Optimal finite-sample experiment design in worst case l/sub 1/ system identification , 1992, [1992] Proceedings of the 31st IEEE Conference on Decision and Control.