No abstract

The paper presents a new big data based control design for autonomous vehicles. The main contribution of this work is the longitudinal velocity optimization process, which is based on the approximation of the reachability sets of a passenger vehicle by using a machine-learning approach. The data, which is used for the approximation, is provided by the high-fidelity car simulation software, CarSim. The approximation is performed by applying a well-known decision tree algorithm, C4.5. The reachability sets are computed for different longitudinal velocities. Moreover, a LPV technique based lateral control design is proposed, which is used to guarantee the trajectory tracking of the vehicle. To enhance the capability of the LPV controller, the control scheme is extended with the longitudinal velocity optimization process. Thus, the stable and safe motion of the vehicle is guaranteed.

This paper presents a novel modeling method for the control design of autonomous vehicle systems. The goal of the method is to provide a control-oriented model in a predefined Linear Parameter Varying (LPV) structure. The scheduling variables of the LPV model through machine-learning-based methods using a big dataset are selected. Moreover, the LPV model parameters through an optimization algorithm are computed, with which accurate fitting on the dataset is achieved. The proposed method is illustrated on the nonlinear modeling of the lateral vehicle dynamics. The resulting LPV-based vehicle model is used for the control design of path following functionality of autonomous vehicles. The effectiveness of the modeling and control design methods through comprehensive simulation examples based on a high-fidelity simulation software are illustrated.

No abstract