Design of a Novel Fuzzy Controller to Enhance Stability of Vehicles

Abstract

This paper presents the design of a novel fuzzy control structure to improve stability of vehicles with semi-active suspension system. The proposed fuzzy controller adjusts the damping coefficient to stabilize the sprung mass and hence reduce the tendency of vehicle to rollover. A full car model with eight degrees of freedom is adopted that includes the vertical, roll, yaw, and pitch motions as well as the vertical motions of each wheel. Four decentralized fuzzy controllers are developed and applied to each individual damper in the vehicle suspension system. The controllers input(s) are lateral acceleration and vehicle states and the output is an adaptive damping coefficient. Mamdani's inference engine is used to obtain the required damping coefficient of each suspension system. To evaluate the performance of the proposed controller, experiments were performed for simple turn and lane change maneuvers. To show the effectiveness of the proposed controller, comparison is made with Cadillac controller. Results show that the fuzzy controller reduces roll angle, linear transfer ration (LTR) and hence decreases the propensity to rollover in vehicles.