Model Predictive Control for Path Following of Autonomous Vehicle Considering Model Parameter Uncertainties

This paper proposes a robust controller for lateral path following on the fundamental of linear-matrix-inequalities-based robust model predictive control (LMI-based RMPC) methodology. Firstly, an uncertain model is established and transformed into polytopic form for path following control, in which norm-bounded uncertain tire corning stiffnesses and time-varying velocity are taken into account. Then, according to the modified polytopic model, the robust controller is designed to provide steering input by solving online minmax optimization for coping with path following problem under aforementioned uncertain factors. Finally, for validating the effectiveness and performance of proposed controller, simulation is carried out and analyzed. The results show that the proposed approach has strong adaptability to severe driving condition with better capabilities on eliminating tracking errors and guaranteeing lateral stability.