Robust Model Predictive Control for Active Suspension System using Linear Matrix Inequalities

In this article, a robust model predictive control (RMPC) using linear matrix inequalities (LMIs) is proposed for vehicle suspension design with parameter uncertainties. Since, in vehicle suspension design, it is desired to improve ride comfort and road holding while satisfying suspension constraints such as suspension deflection and maximum of control input, model predictive control is proposed which is among the most common approaches in constrained optimization problems. On the other hand, to handle suspension constraints, linear matrix inequalities are utilized here. Stability of the designed suspension system is proved, if the proposed linear matrix inequalities are feasible. In addition, uncertain parameters in suspension system are inevitable. In this paper, model predictive control is extended to care for parameter uncertainties by proposing new LMIs. To evaluate the effectiveness of the proposed approach, the proposed control method is applied to quarter car suspension model with parameter uncertainty. Simulation results endorse that the designed controller shows a competitive robust performance while satisfying suspension constraints existing parameter uncertainties. Moreover, simulations with different road profiles, show that the proposed controller is independent from various road excitations.