Designing active vehicle suspension system using critic-based control strategy

Abstract

In this paper, an adaptive critic-based neurofuzzy controller is presented for a 2 DOF active vehicle suspension system with a servo hydraulic actuator. Fuzzy critic-based learning is a reinforcement learning method based on dynamic programming. The only information available for the critic agent is the system feedback, which is interpreted as the last action performed by the controller in the previous state. The signal produced by the critic agent is used alongside the algorithm of error back propagation to tune online conclusion parts of the fuzzy inference rules of the adaptive controller. Simulation results demonstrate the superior performance of this control method in terms of well disturbance rejection, improved ride comfort, robustness to model uncertainty and lower controller cost.