Robust Parameter Estimation

  • I. D. Landau
  • R. Lozano
  • M. M’Saad
Part of the Communications and Control Engineering book series (CCE)


In the previous chapters, we assumed that:
  1. 1.

    The true plant model and the estimated plant model have the same structure (the true plant model is described by a discrete time model with known upper bounds for the degrees n A , n B + d).

  2. 2.

    The disturbances are zero mean and of stochastic nature (with various assumptions).

  3. 3.
    For parameter estimation in closed loop operation, the controller
    1. a.

      has constant parameters and stabilizes the closed loop.

    2. b.

      contains the internal model of the deterministic disturbance for which perfect state disturbance rejection is assured.

  4. 4.

    The parameters are constant or piece-wise constant.

  5. 5.

    The domain of possible parameters values is in general not constrained (exception: recursive maximum likelihood and adaptive filtered closed loop output error).



Closed Loop Plant Model Dead Zone Unmodelled Dynamic Adaptation Error 
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.

Copyright information

© Springer-Verlag London 1998

Authors and Affiliations

  • I. D. Landau
    • 1
  • R. Lozano
    • 2
  • M. M’Saad
    • 3
  1. 1.GR-Automatique EnsiegSt Martin d’HèresFrance
  2. 2.CNRS, HEUDIASYCCompiegne Technological UniversityCompiegne CedexFrance
  3. 3.Caen University, ISMRACaen CedexFrance

Personalised recommendations