부산대학교 도서관

Under particular circumstances, such as changing lanes quickly, traveling on roads with low road adhesion, and so on, it is challenging to guarantee the precision of route tracking and make the vehicle move forward steadily for intelligent vehicles in the meantime. Aiming at this problem, this paper proposed a control strategy combining route tracking and active yaw moment. Firstly, the influence of the prediction horizon of model predictive control (MPC) on vehicle stability is analyzed, and as a steering controller, a model predictive controller with a flexible prediction horizon is created. Then, an active yaw moment controller adopting a form of two-layer structure is designed. The first layer calculates the active yaw moment required by the vehicle. The sideslip angle's effect on the vehicle's stability is taken into account while designing the sliding mode surface, and the weight coefficient varies according to the phases of the sideslip angle phase diagram. The lowest layer aims for the highest tire stability margin, transforms the torque optimal distribution problem into a quadratic programming problem, and obtains the torque of each wheel. As a method for evaluating the effectiveness of the framework and the control algorithm, simulation using CarSim/Simulink is the final step. The outcomes of the simulation test show that the coordinated approach can maintain outstanding vehicle stability under challenging circumstances and has adequate path tracking accuracy.