Title :
Design of observer for networked control systems with non-Gaussian noise
Author :
Li Chao ; Wang Yan
Author_Institution :
Sch. of Autom. Sci. & Electr. Eng., Beihang Univ., Beijing, China
Abstract :
In this paper, a minimum entropy (ME) observer is proposed for nonlinear networked control system (NCS) with non-Gaussian time delay and noise. The observer gain is updated using information theoretic learning (ITL) techniques so that the entropy of the error is minimized. The gradient descent algorithm is applied to get the update rule for the observer gain. Finally, an illustrative example is provided to demonstrate the effectiveness of the proposed approach.
Keywords :
Gaussian noise; closed loop systems; control system synthesis; delays; entropy; feedback; gradient methods; learning (artificial intelligence); networked control systems; nonlinear control systems; observers; ITL techniques; ME observer; NCS; closed-loop feedback control system; gradient descent algorithm; information theoretic learning techniques; minimum entropy observer; nonGaussian noise; nonGaussian time delay; nonlinear networked control system; observer design; observer gain; Adaptive systems; Delay effects; Entropy; Networked control systems; Noise; Observers; ITL; MEE observer; NCS; non-Gaussian; time delay;
Conference_Titel :
Control Conference (CCC), 2013 32nd Chinese
Conference_Location :
Xi´an
