Study on navigation control method for CyberCar based on machine vision

The guiding principle and composition of CyberCar based on machine vision was introduced. Applying IM sequence signals as input response signals and least squares method to establish the dynamic equation for CyberCar steering system by system identification experiments firstly, and then combined with the preview kinematics model and two-degree steering dynamic model of vehicle. Therefore, the steering control mathematics model based on preview kinematics for CyberCar was established. And then the switching hyper plane is designed by applying the optimal control theory, during the change of the curve curvature radius is little, the curve tracking of the intelligent vehicle is carried out by adopting sliding variable structure controller. Aim at the parameter uncertainty of tire, a vehicle steering system uncertain model was founded and analyzed. A H_x optimal controller is designed with the method of H_x control theory to solve the problem about model uncertainty. The simulation and experiment results show that the controller designed by the proposed method has good robustness and adaptability; it can trace the path steadily and reliably.