Integrated Controller Design for Path Following in Autonomous Vehicles

This paper focuses on the problem of path following in autonomous vehicles. A solution to this problem is a strategy during which a feedback/feed-forward control law is considered to control the path of the vehicle. In this controller, feedback and feed-forward gains are calculated based on the vehicle dynamic and road-concerned variables. A linear quadratic regulator (LQR) technique is applied to a three-degree of-freedom (DOF) linear vehicle model and three types of controllers are designed with the use of active front steering (AFS), four-wheel steering (4WS) and combination of an AFS with a direct yaw moment control (DYC) systems. A comparison is made about the performance and the efficacy among the different controllers in two different maneuvers. An 8-DOF nonlinear vehicle model is used for simulation purposes. It is concluded that 4WS system shows better potential for path following control of an autonomous vehicle.