Title :
L1 prediction error approach in system identification
Author :
Carmona, J.C. ; Alvarado, V.M.
Author_Institution :
LSIS, CNRS, Marseille, France
Abstract :
In this paper we present our recent work on model estimation and model validation. Some interesting asymptotic properties of L1 estimated models are settled. As a consequence, a final predictive error criterion in L1-norm is derived. Experimental results on a propagative phenomenon are presented and discussed showing the interest of L1 identification approach.
Keywords :
error analysis; identification; minimisation; state estimation; minimisation; model estimation; model validation; predictive error; robust control; state estimation; system identification; Context modeling; Gaussian distribution; Laplace equations; Large scale integration; Mathematical model; Predictive models; Robustness; Statistical analysis; System identification; Testing;
Conference_Titel :
American Control Conference, 2002. Proceedings of the 2002
Print_ISBN :
0-7803-7298-0
DOI :
10.1109/ACC.2002.1025286
