Title :
Adaptive identification by stochastic approximation
Author :
Vellekoop, Michel ; Bagchi, Arunabha
Author_Institution :
Centre for Process Syst. Eng., Imperial Coll. of Sci., Technol. & Med., London, UK
Abstract :
In this paper a new identifier for continuous-time SISO systems is proposed which differs from a widely studied adaptive scheme only by a time-varying “gain” term, inspired by results in stochastic approximation. We prove that this small modification makes the identifier robust to the presence of noise and along the way we derive a stability result for differential equations driven by diminishingly exciting signals
Keywords :
adaptive estimation; approximation theory; differential equations; identification; stability; adaptive identification; continuous-time SISO systems; differential equations; diminishingly exciting signals; robust identifier; stochastic approximation; time-varying gain term; Approximation algorithms; Convergence; Differential equations; Least squares approximation; Linear systems; Robust stability; Stochastic processes; Stochastic resonance; Stochastic systems; Working environment noise;
Conference_Titel :
Decision and Control, 1995., Proceedings of the 34th IEEE Conference on
Conference_Location :
New Orleans, LA
Print_ISBN :
0-7803-2685-7
DOI :
10.1109/CDC.1995.479202
