LAUSR

Abstract This paper presents a data-driven modeling approach and control design for automated driving purposes. The parameters of the control-oriented polytopic model is tuned using machine-learning algorithm in a Linear Parameter Varying (LPV) structure. The control of the automated driving is designed based on the LPV control synthesis method, with which the performances of the system are guaranteed. Through the automated driving system the steering intervention is performed, while the maximization of the longitudinal velocity in a predicted safety region is achieved. The operation and the effectiveness of the proposed control system is demonstrated through a comprehensive simulation example using the high-fidelity simulation software CarSim.