Fast model predictive control for autonomous vehicle

Abstract

In this paper, we present how model predictive control (MPC) is applied to lane change system based on the vision module. The proposed MPC reduces computing time in finding constrained optimal solutions. For the case of fast systems, real-time solution is very difficult to get in the sample time for the system having complexity MPC formulation or many constraints. Weighted forgetting factors are introduced to find effective active constraints. The proposed fast MPC is quickly computable for not only complex MPC formulation but also in the case of no solution. Performance of the proposed control method is validated via simulations implemented in MATLAB/ Simulink and CarSim in real-time environment.