Iterative Adaptive Control: Windsurfing with Confidence

  • Robert L. Kosut


Learning, apparetly, is the result of making mistakes. A good example is learning to windsurf. The approach I took is simple. Fall Often. This is the approach I seem to use regularly when learning anything. It helped a great deal that I did this in Tahiti where both the air and water were 80 degrees and there was little wind. I was in Tahiti following my first visit with Brian Anderson in Canberra in 1984. A few years later when I was describing my learning experience at windsurfing to Brian, who by the way is an excellent windsurfer, we realized that how a human learns windsurfing might be a procedure that could be used for adaptive control. We speculated about human learning of physical activities as fallows. The human first learns to control over a limited bandwidth and learning pushes out the bandwidth over which an accurate knowledge of the”plant” is known. The human first implements a low-gain controller and learning allows the loop to be tightened. The human tends to “over-react” when the environment changes, often resorting to high- gain or no-gain. Then falls. The few instants just before falling may provide some new information about the next higher increments in the bandwidth of the system. So of course mistakes are fundamental to learning. (This is why stabilizing feedback is negative; the popular concept of positive feedback may be useful, in fact, when it is actually negative!) Our speculations on this “windsurfing” approach to adaptive control was first expressed in [2] and subsequently filled out in a series of papers [24,25].


Adaptive Control Control Design Uncertainty Model Controller Parameter Candidate Controller 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    BDO Anderson, R.R. Bitmead, C.R. Johnson, Jr., P.V. Kokotovic, R.L. Kosut, I.M.Y. Mareels, L. Praly, and B.D. Riedle, (1986), Stability of Adaptive Systems: Passivity and Averaging Analysis, MIT Press, 1986Google Scholar
  2. 2.
    B.D.O.Anderson, and R.L.Kosut, Adaptive Robust Control: On-Line Learning, Proc CDC’91, Brighton, England, Dec 1991.Google Scholar
  3. 3.
    K. J. Äström and B. Wittenmark, Adaptive Control, Addison-Wesley, 1995.Google Scholar
  4. 4.
    K. J. Äström and J. Nilsson, “Analysis of a scheme for iterated identification and control”, Proc. IFAC Symp. on Identification, pp. 171–176, Copenhagen, Denmark, 1994.Google Scholar
  5. 5.
    K. J. Äström, “Matching criteria for control and identification”, Proc. 2nd European Control Conference, pp. 248–251, Groningen, Holland, June 1993.Google Scholar
  6. 6.
    S. Boyd, L. El Ghaoui, E. Feron, and V. Balakrishnan, Linear Matrix Inequalities in System and Control Theory, SIAM Studies in Applied Mathematics, 1994.Google Scholar
  7. 7.
    P. B. Brugarolas, V. Fromion, and M. G. Safonov, “Robust switching missile autopilot”, Proc. 1998 ACC, pp. 3665–3669, Philadelphia, PA, June 24–26, 1998.Google Scholar
  8. 8.
    C. Foias and A. E. Frazho,The Commutant Lifting Approach to Interpolation Problems, Birkhauser-Verlag, 1990Google Scholar
  9. 9.
    M. Gevers, “Towards a joint design of identification and control,”Essays on Control: Perspectives in the Theory and its Applications, editors: ELL. Trentelman and J.C. Willems, pp. 111–151, Birkhauser, Boston, MA, 1993.Google Scholar
  10. 10.
    G.C. Goodwin, M. Gevers, and B. Ninness, “Quantifying the error in estimated transfer functions with applications to model order selection”, IEEE Trans, on Automatic Control, vol. 37, no.7, pp. 913–928, July 1992.MathSciNetMATHCrossRefGoogle Scholar
  11. 11.
    U. Grenander and G. Szebo, Toeplitz Forms and Their Applications, Chelsea Publishing Co., New York, 1958.MATHGoogle Scholar
  12. 12.
    H. Hjalmarsson, S. Gunnarsson, and M. Gevers, “A convergent iterative re¬stricted complexity control design scheme”, Proc. 1994 CDC, pp. 1735–1740, Or¬lando, Florida, Dec. 1994.Google Scholar
  13. 13.
    H. Hjalmarsson, S. Gunnarsson, and M. Gevers, “Optimality and sub- optimality of iterative identification and control design schemes”, Proc. 1995 ACC, pp. 2559–2563, Seattle, Washington, June 1995.Google Scholar
  14. 14.
    L. Kammer, R. Bitmead, P. Bartlett, “Direct iterative tuning via spectral analysis”, Proc. 1998 CDC, pp. 2874–2879, Tampa, FL, Dec. 1998.Google Scholar
  15. 15.
    R. L. Kosut, “Sub-optimal control of linear time-invariant systems subject to control structure constraints”, IEEE Trans, on Auto. Contr., AC-15(5):557–563, June 1970.Google Scholar
  16. 16.
    R. L. Kosut, “Adaptive calibration: An approach to uncertainty modeling and on-line robust control design”, Proc. 25th IEEE CDC, Athens, Greece, Dec. 1986.Google Scholar
  17. 17.
    R.L. Kosut, G.C. Goodwin, and M.P. Polis, editors, “Special Issue on System Identification for Robust Control Design”, IEEE Trans, on Automatic Control, vol. 37, no.7, pp. 899–1008, July 1992.CrossRefGoogle Scholar
  18. 18.
    R. L. Kosut, M. Lau, and S. Boyd, “Set-membership identification of systems with parametric and nonparametric uncertainty”, “Special Issue on Identification for Robust Control Design”, IEEE Trans. Aut Contrvol. 37, no. 7, July, 1992.Google Scholar
  19. 19.
    R.L. Kosut, “Uncertainty model unfalsification: a system identification paradigm compatible with robust control design”, Proc. 1995 CDC, New Orleans, LA, Dec. 1995.Google Scholar
  20. 20.
    R. L. Kosut, “Iterative adaptive robust control via uncertainty model unfalsification”, Proc. 1996 IFAC World Congress, June, 1996, San Francisco, CA.Google Scholar
  21. 21.
    R. L. Kosut and B. D. O. Anderson, “Uncertainty model unfalsification”, Proc. 1997 CDC, San Diego, CA, Dec, 1997.Google Scholar
  22. 22.
    R. L. Kosut, “Uncertainty model unfalsification”, CD-ROM SYSID 2000, Santa Barbara, CA, 21–23 June 2000.Google Scholar
  23. 23.
    M. V. P. Kruger and K. Poolla, “Validation of uncertainty models in the pres¬ence of stochastic disturbance signals”, Selected Topics in Modelling and Control, Vol. 11, pp.1-8, Delft University Press, Dec. 1998Google Scholar
  24. 24.
    W.S. Lee, B.D.O. Anderson, R.L. Kosut, and I.M.Y. Mar eels, “A new approach to adaptive robust control”, Int. J. of Adaptive Control and Signal Processing, vol. 7, pp. 183–211, 1993.MATHCrossRefGoogle Scholar
  25. 25.
    W.S. Lee, B.D.O. Anderson, I.M.Y. Mareels, and R.L. Kosut, “On some key issues in the windsurfer approach to robust adaptive control”, Automatica, Vol. 31, No. 11, pp. 1619–1636, 1995.MathSciNetMATHCrossRefGoogle Scholar
  26. 26.
    M.M. Livestone, M.A. Dahleh, and J.A. Farrell, “A framework for robust control based model invalidation”, Proc. 1994 ACCt pp. 3017–3020, Baltimore, MD, June 1994Google Scholar
  27. 27.
    L. Ljung, System Identification: Theory for the User, First Edition, Prentice- Hall, 1987.MATHGoogle Scholar
  28. 28.
    L. Ljung, “Model validation and model error modeling”, The Astrom Symposium on Control, pp. 15–42, Lund, Sweden, August 27–28, 1999.Google Scholar
  29. 29.
    M. Massoumnia and R. L. Kosut, “A family of norms for system identification problems”, Proc. 1993 ACC, San Francisco, CA, June 1993, and IEEE Trans. Aut Control, vol. 39, no. 5, pp. 1027–1031, May 1994.Google Scholar
  30. 30.
    M. P. Newlin and R. S. Smith, “A generalization of the structured singular value and its application to model validation”, IEEE Thins. Aut Contrvol 43, no. 7, pp. 901–907, July 1998Google Scholar
  31. 31.
    F. Paganini, “A set-based approach for white noise modeling” IEEE Trans. Aut Contr., vol. 41, no. 10, pp. 1453–1465, Oct. 1996.MathSciNetMATHCrossRefGoogle Scholar
  32. 32.
    S. M. Phillips, E. L. Kosut, G. F. Franklin, “An averaging analysis of discrete- time indirect adaptive control”, Proc. 1988 American Control Conference, Atlanta, GA, 1988.Google Scholar
  33. 33.
    K. Poolla, P. Khargonnekar, A. Tikku, J. Krause, and K. Nagpal, “A timedomain approach to model validation”, Proc. 1992 ACC, Chicago, IL, June 1992, and IEEE Trans. Aut Contr., vol 39, no. 5, pp. 951–959, May 1994.Google Scholar
  34. 34.
    M. G. Safonov and T. C. Tsao, “The unfaisified control concept and learning”, IEEE Trans. Aut Contr., vol. 42, no. 6, pp. 843–847, June 1997.MathSciNetMATHCrossRefGoogle Scholar
  35. 35.
    R. S. Smith and J. C. Doyle, “Model validation - a connection between robust control and identification”, Proc. 1989 ACC, pp. 1435–1440, June 1989, and IEEE Trans. Aut Contr., vol. 37, pp. 942–952, July 1992.MathSciNetMATHCrossRefGoogle Scholar
  36. 36.
    B. Wahlberg and P.M. Makila (1996), “On approximation of stable linear dynamical systmes using Laguerre and Kautz functions”, Automatica, 32(5), 693–708.MathSciNetMATHCrossRefGoogle Scholar
  37. 37.
    B. R. Woodley, R. L. Kosut, and J. P. How, “Uncertainty model unfalsification with simulation”, Proc. 1998 ACC, pp. 2754–2755, Philadelphia, PA, June 24–26, 1998.Google Scholar
  38. 38.
    B. R. Woodley, J. P. How, and R. L. Kosut, “Direct unfaisified controller designsolution via convex optimization”, submitted 1999 ACC, San Diego, CA, June 2–4, 1999.Google Scholar
  39. 39.
    Z. Zang, R. R. Bitmead, and M. Gevers, “H2 iterative model refinement and control robustness enhancement”, Proc. 30th IEEE CDC, pp. 279–284, Brighton, UK, Dec. 1991.Google Scholar
  40. 40.
    K. Zhou and J. C. Doyle, Essentials of Robust Control, Prentice Hall, 1998.Google Scholar

Copyright information

© Springer-Verlag London Limited 2001

Authors and Affiliations

  • Robert L. Kosut
    • 1
    • 2
  1. 1.SC Solutions, IncSanta ClaraUSA
  2. 2.Applied Computation & Mathematics Program, under NASA Grant No. NAG-1-1964Research supported by DARPAUSA

Personalised recommendations