ECE 417 Identification & Adaptive Control
 
This is a graduate-level course on the foundations of identification and adaptive control and learning in control systems. Various algorithms will be developed for identification and control of linear and nonlinear systems. The course emphasizes analytical techniques for verifying stability and evaluating performance, and provides a broad set of tools for control of complex
systems.To follow this course, some background in control theory fundamentals
at the level of ECE 415 is required. Some familiarity with stochastic
processes will be very helpful, although all of the necessary concepts will
be reviewed in class.

Text: The following are useful, but not required:

  • Stochastic systems : estimation, identification, and adaptive control, P.R. Kumar, P. Varaiya. Prentice Hall, 1986.
  • Adaptive filtering prediction and control, G. C. Goodwin and K-S. Sin. Prentice-Hall, 1984.
  • Adaptive control, 2nd ed., K. Astrom and B. Wittenmark, Addison-Wesley, 1995.
  • System identification: theory for the user, L. Ljung. Prentice-Hall, 1987.
  • Identification and stochastic adaptive control, H-F. Chen and L. Guo, Birkhauser, 1991.
  • Probability, Random Processes, And Estimation Theory For Engineers, H. Stark and J.W. Woods, second edition, Prentice Hall, 1994.

Lecture notes are available at TIS on Green Street

Brief Outline

I. Identification

  • Identification & prediction
  • Introduction to algorithms
  • ARMA models
  • Least squares & related algorithms

II. Averaging analysis

  • White noise & spectral factorization
  • ELS & IV methods
  • The ODE method
  • Stability verification

III. Adaptive prediction & control

  • Self-tuning regulators
  • Delay compensation
  • Model reference control