Search Results

This paper describes an automated path following system using vision sensor. Lateral control law for path following is especially underlined which is developed by using the backstepping control design methodology. To establish the proposed control system, the lateral offset to the reference path, the heading angle of vehicle relative to tangent line to the path, and path curvature are required. Those inputs to the controller have been calculated through Kalman filter which is frequently adopted for the purpose. The lane mark detection has been achieved in an ECU (Electric Control Unit) platform with vision sensor. The yaw rate and side-slip angle also needed in the controller are estimated by Kalman estimator. To show the performance of the proposed controller under different speeds, experiment has been conducted on a proving ground having straight and curve sections with the curvature of about 260m.

This paper mainly focuses on a lateral control law for pre-given path following which is developed by using the backstepping control design methodology. The position information of the vehicle is obtained by Real Time Kinematic DGPS, and the yaw rate and side-slip angle used in controller are estimated by Kalman estimator. To show the performance of the proposed controller under different speed and various path curvature conditions, the results are given through experiments which are executed on proving ground especially designed for high maneuvering test of which minimum radius of curvature is about 60 m.