Title :
Adaptive identifier with a self-tuning filter
Author_Institution :
Dept. of Electr. & Comput. Eng., Iowa Univ., IA, USA
Abstract :
It is shown that the performance of the adaptive identifier of equation error type can be improved substantially by updating the filter online. It is known that, in the presence of uncertainty, the equation error method would not produce an estimate which has a good fit at low frequencies. To fix the problem, the identification schemes are modified by making filters time varying, and these time varying filters are updated online such that the estimate converges to a value that would be obtained by the output error method. In this way, the effects of the uncertainty are reduced
Keywords :
adaptive control; filtering and prediction theory; identification; self-adjusting systems; adaptive control; adaptive identifier; identification; self-tuning filter; time varying filters; uncertainty; Adaptive filters; Cities and towns; Convergence; Equations; Frequency estimation; Knowledge engineering; Parameter estimation; Stability; Transfer functions; Uncertainty;
Conference_Titel :
Decision and Control, 1990., Proceedings of the 29th IEEE Conference on
Conference_Location :
Honolulu, HI
DOI :
10.1109/CDC.1990.203391