MPC-based Path Following Design for Automated Vehicles with Rear Wheel Steering

Abstract

Many studies have been recently exploited to discuss the path following control algorithms for automated vehicles using various control techniques. However, path following algorithm considering the possibility of automated vehicles with rear wheel steering (RWS) is still less investigated. In this study, we implemented nonlinear model predictive control (NMPC) on a passenger vehicle with active RWS for path following. The controller was compared to two other variations of NMPC where the rear steering angle is proportional to the front or fixed to zero. Simulation results suggested that the proposed controller outperforms the other two variations and the baseline controllers (Stanley and LQR) in terms of accuracy and responsiveness.