Optimal composite nonlinear feedback control with multi objective algorithms for active front steering system

Abstract

The main purpose of controlling vehicle handling is to ensure that the vehicle follows the desired path. Vehicle yaw rate must be controlled in order to achieve a good vehicle handling. In this thesis, optimal Composite Nonlinear Feedback (CNF) controller with multi objective algorithms is proposed for the Active Front Steering (AFS) system in improving the vehicle yaw rate response. The model used to validate the performance of the controller is a 7 degree-of-freedom (DOF) nonlinear vehicle model. This vehicle model is also simplified to a 2 DOF bicycle model for the purpose of controller design. In designing the optimal CNF control, the parameter selection of optimal linear and non-linear gain parameters becomes very important to obtain a good system response. Optimization algorithms are utilized to minimize the complexity in selecting the best parameters. Hence, Multi Objective Particle Swarm Optimization (MOPSO) and Multi Objective Genetic Algorithm (MOGA) are proposed to produce the optimal CNF. Moreover, manual tuning method was utilized and has been compared with the proposed algorithms. As a result, the performance of the yaw rate response is improved with a 98 percent reduction in error. Hence, the vehicle handling can be improved and the vehicle will be able to travel safely on the desired path.