» Download Article

Authors
Michel Gevers; Xavier Bombois

Identifier
10.3182/20060329-3-AU-2901.00215

Index Terms
identification,optimal experiment design,optimization,input signals control accuracy

Abstract
In this paper we briefly review the evolution of the main tools and results for optimal experiment design for system identification. The initial work dates back to the seventies and focused on the accuracy of the parameters of the input-output transfer function estimate. In the eighties, new formulas for the variance of transfer function estimates based on high-order model approximations led to the first goaloriented experiment design results. The recent trend is to address control-oriented optimal design questions using the more accurate parameter covariance formulas for finite order models.

References

[1] Barenthin, Märta and Håkan Hjalmarsson (2005).
    Identification and control: joint input design
    and H∞ state feedback with ellipsoidal parametric
     uncertainty. submitted for publication.
[2] Barenthin, Märta, Henrik Jansson and Håkan
    Hjalmarsson (2005). Applications of mixed
    H2 and H∞ input design in identification.
    In: 16th IFAC World Congress on Automatic
    Control, paper 03882. Prague.
[3] Bombois, X. and M. Gilson (2006). Cheapest identification
     experiment with guaranteed accuracy
     in the presence of undermodeling. In:
    submitted to SYSID 2006. Newcastle, Australia.
    
[4] Bombois, X., G. Scorletti, Gevers, P. M. J. Van den
    Hof and R. Hildebrand (2005). Least costly
    identification experiment for control. submittted
     to Automatica.
[5] Bombois, X., G. Scorletti, M. Gevers, R. Hildebrand
     and P. M. J. Van den Hof (2004a).
    Cheapest open-loop identification for control.
    In: CD-ROM Proc. 33rd IEEE Conf. on Decision
     and Control. The Bahamas. pp. 382-387.
[6] Bombois, X., G. Scorletti, P. M. J. Van den Hof and
    M. Gevers (2004b). Least costly identification
    experiment for control: a solution based on
    a high-order model approximation. In: CD-ROM
     Proc. American Control Conference.
    Boston, MA, USA. pp. 2818-2823.
[7] Bombois, X., M. Gevers, G. Scorletti and
    B. D. O. Anderson (2001). Robustness analysis
     tools for an uncertainty set obtained
    by prediction error identification. Automatica
    37(10), 1629-1636.
[8] Boyd, S., L. El Ghaoui, E. Feron and V. Balakrishnan
     (1994). Linear Matrix Inequalities in
    System and Control Theory. SIAM Studies in
    Applied Mathematics. Philadelphia.
[9] Cooley, Brian L. and Jay H. Lee (2001). Control-relevant
     experiment design for multivariable
    systems described by expansions in orthonormal
     bases. Automatica 37(2), 273-281.
[10] Forssell, U. and L. Ljung (2000). Some results
     on optimal experiment design. Automatica
     36, 749-756.
[11] Gevers, M. and L. Ljung (1986). Optimal experiment
      designs with respect to the intended
     model application. Automatica 22, 543-554.
[12] Gevers, M., X. Bombois, B. Codrons, G. Scorletti
     and B. D. O. Anderson (2003). Model validation
      for control and controller validation in
     a prediction error identification framework -
     Part I: theory. Automatica 39(3), 403-415.
[13] Goodwin, G. and R. L. Payne (1977). Dynamic
     System Identification: Experiment Design
     and Data Analysis. Academic Press. New
     York.
[14] Goodwin, G. C., J. S. Welsh and A. Feuer (2005).
     Robust optimal experiment design. Technical
      report. Dep. of Electrical Engineering and
     Computer Science. The University of Newcastle,
      Australia 2308.
[15] Hildebrand, R. and M. Gevers (2003). Identification
      for control: optimal input design with
     respect to a worst-case ν-gap cost function.
     SIAM Journal on Control and Optimization
     41(5), 1586-1608.
[16] Hjalmarsson, H. (2005). From experiment design
     to closed-loop control. Automatica 41, 393-
     438.
[17] Hjalmarsson, H. and H. Jansson (2003). Using a
     sufficient condition to analyze the interplay
     between identification and control. In: CD-ROM
      of 13th IFAC Symposium on System
     Identification. Rotterdam, The Netherlands.
     pp. 45-50.
[18] Hjalmarsson, H., M. Gevers and F. De Bruyne
     (1996). For model-based control design,
     closed-loop identification gives better performance.
      Automatica 32, 1659-1673.
[19] Jansson, H. (2004). Experiment design with applications
      in identification for control. PhD
     thesis. Royal Institute of Technology (KTH).
     TRITA-S3-REG-0404.
[20] Jansson, H. and H. Hjalmarsson (2004a). Convex
     computation of worst-case criteria with applications
      in identification and control. In: IEEE
     Conference on Decision and Control. The Bahamas.
      pp. 3132-3137.
[21] Jansson, H. and H. Hjalmarsson (2004b). A general
      framework for mixed H∞ and H2 input
      design. submitted for publication to IEEE
     Trans. Auto. Control.
[22] Jansson, H. and H. Hjalmarsson (2004c). Mixed
     H∞ and H2 input design for identification. In:
     CD-ROM Proc. 43rd IEEE Conf. on Decision
     and Control. The Bahamas. pp. 388-393.
[23] Jansson, H. and H. Hjalmarsson (2005). Optimal
     experiment design in closed loop. In: 16th
     IFAC World Congress on Automatic Control,
     paper 04528.
[24] Karlin, S. and W. Studden (1966). Tchebycheff
      systems with applications in analysis
     and statistics. Pure Appl. Math. 15, Wiley-Interscience.
      New York.
[25] Lindqvist, K. and H. Hjalmarsson (2001). Identification
      for control: adaptive input design using
      convex optimization. In: CD-ROM Proc.
     of 40th IEEE Conf. on Decision and Control.
     Orlando, Florida.
[26] Ljung, L. (1985). Asymptotic variance expressions
      for identified black-box transfer function
     models. IEEE Trans. Automatic Control AC-
     30, 834-844.
[27] Ljung, L. (1999). System Identification: Theory
     for the User, 2nd Edition. Prentice-Hall. Englewood
      Cliffs, NJ.
[28] Mehra, R. K. (1974). Optimal input signals for parameter
      estimation in dynamic systems - survey
      and new results. IEEE Trans. on Automatic
      Control AC-19(6), 753-768.
[29] Nesterov, Y. and A. Nemirovskii (1994). Interior
     Point polynomial algorithms in convex programming
     . Vol. 13. SIAM Studies in Applied
     Mathematics. Philadelphia.
[30] Ng, T. S., G. C. Goodwin and R. L. Payne (1977a).
     On maximal accuracy estimation with output
     power constraints. IEEE Trans. Automatic
     Control 22, 133-134.
[31] Ng, T. S., G. C. Goodwin and T. Söderström
     (1977b). Optimal experiment design for linear
     systems with input-output constraints. Automatica
      13, 571-577.
[32] Ninness, B. and H. Hjalmarsson (2004). Variance
     error quantifications that are exact for finite
     model order. IEEE Transactions on Automatic
      Control 49(8), 1275-1291.
[33] Rivera, D. E., H. Lee, M. W. Braun and H. D. Mittelmann
      (2003). "plant-friendly" system identification:
      a challenge for the process industries.
      In: Proc. 13th IFAC Symp. on System
     Identification. Rotterdam, The Netherlands.
     pp. 917-922.
[34] Vinnicombe, G. (1993). Frequency domain uncertainty
      and the graph topology. IEEE Trans.
     Automatic Control AC-38, 1371-1383.
[35] Wahlberg, B. and L. Ljung (1986). Design variables
      for bias distribution in transfer function
     estimation. IEEE Trans. Automatic Control
     AC-31, 134-144.
[36] Xie, L. L. and L. Ljung (2001). Asymptotic variance
      expressions for estimated frequency
     functions. IEEE Transactions on Automatic
     Control AC-46, 1887-1899.
[37] Zarrop, M. (1979). Optimal experiment design for
     dynamic system identification. Lecture Notes
     in Control and Information Sciences, Vol. 21,
     Springer Verlag. Berlin, New York.