Abstract
This paper presents a lazy learnig-based neural network controller for a semi-active vehicle suspension system. The proposed controller allows the semi-active suspension system to improve both the ride comfort and the road isolation. The neural-based controller was studied using a quarter-car model. Simulation results show that the neural approach provides a simple and effective approach to the adaptive control of a suspension system. Other advantages this algorithm presents are that it is easy to select the network structure and it learns quickly.
KEYWORDS: Neural networks, semi-active vehicle suspension, lazy learning.