Statistical Learning Control of Uncertain Systems : It is Better Than It Seems

Recently, probabilistic methods and statistical learning theory have been shown to provide approximate solutions to \diÆcult" control problems. Unfortunately, the number of samples required in order to guarantee stringent performance levels may be prohibitively large. This paper introduces bootstrap learning methods and the concept of stopping times to drastically reduce the bound on the number of samples required to achieve a performance level. We then apply these results to obtain more eÆcient algorithms which probabilistically guarantee stability and robustness levels when designing controllers for uncertain systems.

[1]  H. Chernoff.  A Measure of Asymptotic Efficiency for Tests of a Hypothesis Based on the sum of Observations , 1952 .

[2]  Vladimir Vapnik,et al.  Chervonenkis: On the uniform convergence of relative frequencies of events to their probabilities , 1971 .

[3]  B. Anderson,et al.  Output feedback stabilization and related problems-solution via decision methods , 1975 .

[4]  R. Dudley.  Central Limit Theorems for Empirical Measures , 1978 .

[5]  G. Stein,et al.  Multivariable feedback design: Concepts for a classical/modern synthesis , 1981 .

[6]  V. Milman,et al.  Asymptotic Theory Of Finite Dimensional Normed Spaces , 1986 .

[7]  Wansoo T. Rhee,et al.  Martingale Inequalities and NP-Complete Problems , 1987, Math. Oper. Res..

[8]  R. Stengel,et al.  Technical notes and correspondence: Stochastic robustness of linear time-invariant control systems , 1991 .

[9]  George E. Collins,et al.  Partial Cylindrical Algebraic Decomposition for Quantifier Elimination , 1991, J. Symb. Comput..

[10]  Hoon Hong,et al.  Simple solution formula construction in cylindrical algebraic decomposition based quantifier elimination , 1992, ISSAC '92.

[11]  G. Coxson,et al.  Computational complexity of robust stability and regularity in families of linear systems , 1993 .

[12]  Arkadi Nemirovski,et al.  Several NP-hard problems arising in robust stability analysis , 1993, Math. Control. Signals Syst..

[13]  Robert F. Stengel,et al.  A monte carlo approach to the analysis of control system robustness , 1993, Autom..

[14]  P. Dorato,et al.  Static output feedback: a survey , 1994, Proceedings of 1994 33rd IEEE Conference on Decision and Control.

[15]  Kameshwar Poolla,et al.  Statistical validation for uncertainty models , 1995 .

[16]  Michael G. Safonov,et al.  The unfalsified control concept: A direct path from experiment to controller , 1995 .

[17]  Yun Huang,et al.  Soft vs. hard bounds in probabilistic robustness analysis , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

[18]  Pramod P. Khargonekar,et al.  Randomized algorithms for robust control analysis and synthesis have polynomial complexity , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

[19]  R. Tempo,et al.  Probabilistic robustness analysis: explicit bounds for the minimum number of samples , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

[20]  M. Talagrand.  A new look at independence , 1996 .

[21]  Kemin Zhou,et al.  On the probabilistic characterization of model uncertainty and robustness , 1997, Proceedings of the 36th IEEE Conference on Decision and Control.

[22]  Kemin Zhou,et al.  A probabilistic approach to robust control , 1997, Proceedings of the 36th IEEE Conference on Decision and Control.

[23]  Wei Yang,et al.  Robust Multi-Objective Feedback Design by Quantifier Elimination , 1997, J. Symb. Comput..

[24]  Shankar P. Bhattacharyya,et al.  Robust, fragile or optimal? , 1997, Proceedings of the 1997 American Control Conference (Cat. No.97CH36041).

[25]  Marek Karpinski,et al.  Polynomial Bounds for VC Dimension of Sigmoidal and General Pfaffian Neural Networks , 1997, J. Comput. Syst. Sci..

[26]  Kemin Zhou,et al.  Constrained optimal synthesis and robustness analysis by randomized algorithms , 1998, Proceedings of the 1998 American Control Conference. ACC (IEEE Cat. No.98CH36207).

[27]  I. Mareels,et al.  Dead beat controllability of polynomial systems: symbolic computation approaches , 1998, IEEE Trans. Autom. Control..

[28]  Dennis S. Bernstein,et al.  Fixed‐structure robust controller synthesis via decentralized static output feedback , 1998 .

[29]  Mathukumalli Vidyasagar,et al.  Statistical learning theory and randomized algorithms for control , 1998 .

[30]  Roberto Tempo,et al.  Probabilistic Robustness Analysis and Design of Uncertain Systems , 1999 .