Research on System Design and Control Technology of Vision-Based CyberCar

In this paper the vehicle structure, performance parameters and dynamic model of JLUIV5-CyberCar are discussed. An optimal controller is designed to achieve the line path tracking ability, while the curve path tracking ability is guaranteed by a fuzzy controller. To improve the tracking performance of the vehicle steering system in different conditions, controller output collections of JLUIV5-CyberCar are established combined with variable structure control theory. Design the state observer based on Kalman filter theory to estimate the lateral velocity and yaw rate. Simulations and experimental results have confirmed the effectiveness of the proposed method.