Driver steering model based on globally optimal model predictive control

This paper presents a new driver model based on model predictive control (MPC). The proposed driver model is formulated according to four assumptions. The input of the driver model is generalized as a region of the road excluding obstacles. The proposed model has multiple objectives, consisting of reducing driving effort, and ensuring smoothness and safety margins, taking into account predicted future states. The steering wheel angle of the driver is calculated by evaluating whether the objectives have been achieved using the framework of MPC. Moreover, the evaluation is formulated to ensure that a global optimal solution is obtained. The proposed model explains simulation results on a slalom course and performance when passing through a narrow passage.