Title :
Genetic algorithms in system identification
Author :
Kristinsson, K. ; Dumont, G.A.
Author_Institution :
Dept. of Electr. Eng., British Columbia Univ., Vancouver, BC, Canada
Abstract :
Current online identification techniques are recursive and local search techniques. In the present work, it is shown how genetic algorithms, a parallel, global search technique emulating natural genetic operators, can be used to estimate the poles and zeros of a dynamical system. An adaptive controller is designed on the basis of the estimates. Simulations and an experiment show the technique to be satisfactory and to provide unbiased estimates in the presence of colored noise
Keywords :
adaptive control; identification; poles and zeros; adaptive controller; dynamical system; genetic algorithms; global parallel search; local search; poles; recursive search; system identification; zeros; Adaptive control; Biological cells; Colored noise; Genetic algorithms; Genetic mutations; Maximum likelihood estimation; Poles and zeros; Programmable control; Sampling methods; System identification;
Conference_Titel :
Intelligent Control, 1988. Proceedings., IEEE International Symposium on
Conference_Location :
Arlington, VA
Print_ISBN :
0-8186-2012-9
DOI :
10.1109/ISIC.1988.65498
