Modeling and Design of Traction Control System of vehicle using Nonlinear Predictive Control and Neural Network

Traction control system (TCS) is one of the necessary systems for increasing vehicle safety. This control system is used to prevent excessive slipping of wheels especially when the vehicle suddenly starts to move. Keeping the wheels slip in a desirable range under unfavorable weather condition is a challenging issue due to unknown effects of road surface and severe nonlinear behavior of tire during the acceleration process. On the other hand, in designing a controller, the existence of some unknown uncertainties such as un-model dynamics and variation of vehicle parameters should be considered. Therefore, the presence of a nonlinear robust control law seems avoidable for TCS. In this paper, at first, using nonlinear predictive control method, a modern nonlinear optimal controller is designed for TCS. Then, unknown uncertainties of the system are adaptively estimated using a radial basis function neural network (RBFNN). Finally, some simulation results are presented for tracking the reference wheel slip in the presence of uncertainties for different maneuvers in order to assess the behavior of the proposed control system. The results show the effectiveness of the proposed control system against the nonlinear effects and uncertainties.