Yaw Angle Control for Autonomous Vehicle Using Kalman Filter Based Disturbance Observer

Yaw angle control is an essential function in every autonomous vehicle system. In this paper, a novel yaw angle control method for vehicle is proposed. Because the attitude of vehicle obtained from on-board GPS receiver is not only delayed but also at low rate in comparison with the control frequency of EPS servo drives, it is impossible to use GPS directly in yaw angle control loop. By integrating a single antenna GPS receiver with a yaw rate sensor using proposed Kalman filter, GPS measurement delay is handled and yaw angle is estimated at high rate. In order to improve the robustness of yaw angle control, disturbance observer utilizing estimated yaw angle and nominal yaw dynamics is designed. The disturbance observer serves as the inner-loop of the outer one with feedback-feed forward controllers. Simulations and experiments are performed to verify the proposed control system.