Naslov
Neural Network Based Sliding Mode Control of Electronic Throttle

Autori
Barić, Miroslav ; Petrović, Ivan ; Perić, Nedjeljko

Izvornik
Engineering applications of artificial intelligence (0952-1976) 18 (2005); 951-961

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni

Ključne riječi
electronic throttle control; uncertainty; sliding mode; neural networks; adaptive control

Sažetak
In this paper a neural network based sliding mode controller for electronic throttle is proposed. Electronic throttle is considered as an uncertain linear system. The uncertainties, which consist of an unknown friction and spring torque are estimated by the neural network whose parameters are adapted in an on-line fashion. Control and adaptive laws guarantee the boundedness of all signals in the system. Presented experimental results demonstrate the efficiency and robustness of the proposed control scheme.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika

POVEZANOST RADA