Advertisement

Automation and Remote Control

, Volume 80, Issue 9, pp 1628–1644 | Cite as

Robust Identification under Correlated and Non-Gaussian Noises: WMLLM Procedure

  • A. S. PoznyakEmail author
Topical Issue
  • 7 Downloads

Abstract

The paper represents the main ideas of Robust Identification Theory, which was initiated by Ya.Z. Tzypkin in 80s of the past century. Here we demonstrate that the parallel application of both the “Whitening Procedure” and the recurrent min-max version of the “Maximum Likelihood Method” (MLLM), guarantees the property of Asymptotic Consistency of this procedure. The information Cramér–Rao inequality is also obtained and it is shown that this combined procedure attains this information bound. This means that for wide class of regular observations of ARX (auto regression with exogenous input) models, there does not exist any other identification algorithm, estimating asymptotically unknown parameters “quicker” then the procedure discussed here, if the distribution of the white noise at the input of the forming filter is known exactly. The main specific feature of this method consists in the possible consideration of external non-Gaussian white noise (defined on the given class of admissible distributions) in the input of the forming filter, making the noise sequence, affecting the input of ARX-dynamics, non-Gaussian, and correlated. The almost sure convergence as well as the Asymptotic Normality of the estimation error are formulated. If the information on the input of the forming filter is uncertain, that is, when the distribution belongs to a given class, then the Huber’s approach is applied using the robust version of MLLM.

Keywords

robust identification whitening procedure recurrent ML non-Gaussian white noise Cramér–Rao inequality 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Reinelt, W., Garulli, A., and Ljung, L., Comparing Different Approaches to Model Error Modelling in Robust Identification, Automatica, 2002, vol. 38, pp. 787–803.CrossRefzbMATHGoogle Scholar
  2. 2.
    Kosut, R.L., Goodwin, G.C., and Polis, M.P., Eds., Special Issue on System Identification for Robust Control Design, IEEE Trans. Automat. Control, 1992, vol. 37, no. 7, pp. 899–1008.Google Scholar
  3. 3.
    Smith, R.S. and Dahleh, M., Proc. 1992 Santa Barbara Workshop “The Modeling of Uncertainty in Control Systems.” Lect. Notes Control and Inform. Sci., vol. 192, New York: Springer-Verlag, 1994.Google Scholar
  4. 4.
    Söderström, T. and Aström, K.J., Editorial. Special Issue on Trends in System Identification, Automatica, IFAC, 1995, P. 1689–1690.Google Scholar
  5. 5.
    Garulli, A., Tesi, A., and Vicino, A., Eds., Robustness in Identification and Control, Lect. Notes Control and Inform. Sci., vol. 245, Berlin: Springer, 1999.Google Scholar
  6. 6.
    Goodwin G.C. Identification and Robust Control: Bridging the Gap, Proc. 7th IEEE Mediterranian Conf. on Control and Automation, 1999, Israel, Haifa.Google Scholar
  7. 7.
    Goodwin, G.C., Braslavsky, J.H., and Seron, M.M., Non-Stationary Stochastic Embedding for Transfer Function Estimation, Proc. 14th IFAC World Congr., 1999, China, Bejing.Google Scholar
  8. 8.
    Ljung, L., Model Validation and Model Error Modeling, in The Astrom Symposium on Control, Lund, Sweden, August 27–28, 1999, pp. 15–42.Google Scholar
  9. 9.
    Ljung, L., System Identification—Theory for the User, Upper Saddle River: Prentice Hall, 1999, 2nd ed.zbMATHGoogle Scholar
  10. 10.
    Walter, E. and Piet-Lahanier, H., Estimation of Parameter Bounds from Bounded-Error Data: A Survey, Math. Comput. Simulat., 1990, vol. 32, pp. 449–468.MathSciNetCrossRefGoogle Scholar
  11. 11.
    Wahlberg, B. and Ljung, L., Hard Frequency-Domain Model Error Bounds from Least-Squares Like Identification Techniques, IEEE Trans. Automat. Control, 1992, vol. 37, no. 7, pp. 900–912.MathSciNetCrossRefzbMATHGoogle Scholar
  12. 12.
    Giarré, L., Milanese, M., and Taragna, M., H Identification and Model Quality Evaluation, IEEE Trans. Automat. Control, 1997, vol. 42, no. 2, pp. 188–199.MathSciNetCrossRefzbMATHGoogle Scholar
  13. 13.
    Garulli, A., Vicino, A., and Zappa, G., Conditional Central Algorithms for Worst-Case Set Membership Identification and Filtering, IEEE Trans. Automat. Control, 2000, vol. 45, no. 1, pp. 14–23.MathSciNetCrossRefzbMATHGoogle Scholar
  14. 14.
    Polyak, B.T. and Tsypkin, Ya.Z., Robust Estimation under Incomplete Information, in Cybernetics Problems, Adaptive Control Systems, Moscow: Akad. Nauk SSSR, 1977, pp. 6–15.Google Scholar
  15. 15.
    Polyak, B.T. and Tsypkin, Ya.Z., Adaptive Estimation Algorithms: Convergence, Optimality, Stability, Autom. Remote Control, 1979, vol. 40, no. 3, pp. 378–389.MathSciNetzbMATHGoogle Scholar
  16. 16.
    Polyak, B.T. and Tsypkin, Ya.Z., Robust Identification, Automatica, 1980, vol. 16, no. 1, pp. 53–63.MathSciNetCrossRefzbMATHGoogle Scholar
  17. 17.
    Tsypkin, Ya.Z. and Poznyak, A.S., Recurrent Optimization Algorithms in Uncertainties, Itogi Nauki Tekh., Ser. Tekh. Kibernetika, 1983, vol. 16, pp. 3–69.Google Scholar
  18. 18.
    Tsypkin, Ya.Z., Poznyak, A.S., and Tikhonov, S.N., Optimal Methods of Adaptive Identification, Itogi Nauki Tekh., Ser. Tekh. Kibern., 1990, vol 29, pp. 3–44.MathSciNetGoogle Scholar
  19. 19.
    Poznyak, A.S., Recursive Stochastic Gradient Procedures in the Presence of Dependent Noise, Lect. Notes Control and Inform. Sci. IIASA 81. Stochastic Optimization, Springer-Verlag, 1984, pp. 522–533.Google Scholar
  20. 20.
    Poznyak, A.S., Estimating the Parameters of Autoregressing Processes by the Method of Least Squares, Int. J. Systems Sci., 1980, vol. 5, no. 11, pp. 235–254.MathSciNetGoogle Scholar
  21. 21.
    Poznyak, A.S. and Chikin, D.O., Asymptotical Properties of Stochastic Approximation Procedures with Dependent Noises, Autom. Remote Control, 1984, vol. 45, no. 12, pp. 1601–1615.zbMATHGoogle Scholar
  22. 22.
    Chikin, D.O. and Poznyak, A.S., Gradient Procedures of Stochastic Approximation with Dependent Noise and their Asymptotic Behavior, Int. J. Systems Sci., 1985, vol. 16, no. 8, pp. 917–949.CrossRefGoogle Scholar
  23. 23.
    Poznyak, A.S. and Tsypkin, Ya.Z., Optimization Algorithms Optimal on a Class in the Presence of Correlated Noise, USSR Comput. Math. Math. Phys., 1984, vol. 24, no. 3, pp. 112–122.MathSciNetCrossRefzbMATHGoogle Scholar
  24. 24.
    Tsypkin, Ya.Z. and Poznyak, A.S., Generalized Method of Instrumental Variables in Problems of Identification of Linear Objects, Dokl. Akad. Nauk USSR, 1989, vol. 306, pp. 1068–1073.MathSciNetGoogle Scholar
  25. 25.
    Poznyak, A.S. and Tikhonov, S.N., Strong Consistency of Nonlinear Recursive Algorithms of Estimation of the Parameters of Linear Difference Equations, Autom. Remote Control, 1990, vol. 51, no. 6, pp. 798–807.zbMATHGoogle Scholar
  26. 26.
    Poznyak, A.S. and Tikhonov, S.N., On Extension of Nonlinear Residual Transformation Class for Recursive Algorithms of Parametric Estimation, Autom. Remote Control, 1990, vol. 51, no. 10, pp. 1418–1424.zbMATHGoogle Scholar
  27. 27.
    Poznyak, A.S. and Tikhonov, S.N., Asymptotic Normality and the Rate of Convergence of Recursive Stochastic Processes with a Nonlinear Residual Transformation, Autom. Remote Control, 1992, vol. 53, no. 8, pp. 1222–1230.zbMATHGoogle Scholar
  28. 28.
    Bondarenko, M.V. and Poznyak, A.S., Asymptotic Normality and an Estimate for the Rate of Convergence of Algorithms for the Identification of Nonstationary Objects, Autom. Remote Control, 1992, vol. 53, no. 7, pp. 989–999.MathSciNetGoogle Scholar
  29. 29.
    Kaz’min, A.V. and Poznyak, A.S., Recurrent Estimation of the Parameters of Autoregression-Exogenous Input Models with Noise Described by Autoregression-Moving Average Processes, Autom. Remote Control, 1992, vol. 53, no. 10, pp. 1549–1556.MathSciNetGoogle Scholar
  30. 30.
    Bondarenko, M.V. and Poznyak, A.S., Convergence of Algorithms for Estimating the Nonstationary Parameters of Regression-Autoregression Plants under Noise of Moving Averages Type, Autom. Remote Control, 1993, vol. 54, no. 8, pp. 1255–1271.MathSciNetGoogle Scholar
  31. 31.
    Platonov, A.A., Poznyak, A.S., Tikhonov, S.N., and Shabatin, E., Analysis of Generalized Accelerated Identification Algorithms, Autom. Remote Control, 1993, vol. 54, no. 2, pp. 311–322.MathSciNetzbMATHGoogle Scholar
  32. 32.
    Poznyak, A.S. and Tikhonov, S.N., Strong Consistency of the Extended Least Squares Method with Nonlinear Error Transformation, Autom. Remote Control, 1990, vol. 51, no. 8, pp. 1105–1112.zbMATHGoogle Scholar
  33. 33.
    Huber, P., Robustness and Design, in A Survey of Statistical Design and Linear Models, North-Holland Publisher Company, 1975.Google Scholar
  34. 34.
    Poznyak, A.S., Advanced Mathematical Tools for Automatic Control Engineers. Vol. 2. Stochastic Techniques, New York: Elsevier, 2009.Google Scholar

Copyright information

© Pleiades Publishing, Ltd. 2019

Authors and Affiliations

  1. 1.CINVESTAV-IPNMexico CityMexico

Personalised recommendations