Research on Adaptive Model Prediction of Lane Keeping Control Based on Different Grades of Roads

Due to the different curvatures of intelligent vehicles on roads of different levels, it is difficult to ensure the safety and stability of vehicles on roads of different levels. In response to the above issues, this article constructs a vehicle kinematic model and, based on the corresponding lane keeping sensor fusion, applies adaptive model prediction theory to establish different levels of road models in Carsim. Combining the real-time curvature of different levels of roads and the constraint conditions of road speed, a linear time invariant controller is designed with lateral position deviation and relative yaw rate deviation as control variables. Design a controller model in Simulink and compare the simulation results with Carsim's built-in driver model. The simulation results show that the controller designed by the adaptive model predictive control theory can effectively improve the safety and comfort of vehicle obstacle avoidance.