Prediction-based controller radial neural network for the traction control system

One of the systems needed to improve vehicle safety is the traction control system (TCS). This control mechanism keeps the wheels from slipping too much when the car accelerates, especially when it moves quickly. Because of the significant nonlinear behavior of the tire during acceleration and the unknown impacts of the road surface, it might be difficult to maintain wheel slip in an acceptable range, especially in bad weather. However, some uncertainties, such as unmodeled dynamics and vehicle parameter uncertainty, should be taken into account when designing the controller. Consequently, TCS requires the existence of a strong nonlinear control law. In this study, an analytical design for a TCS controller is made using the method of nonlinear predictive control. The control system's resistance is then increased by employing an adaptive radial basis neural network (RFNN) to predict the system's unknown uncertainties. In this study, the controller was designed using half car and quarter car models, respectively. The behavior of the suggested control system for tracking the reference wheel's slip in the face of uncertainty for various movements is examined in the simulation results that follow. The resulting results have been compared with the simulation results generated from the nonlinear sliding mode controller response used in valid articles in order to provide a more thorough examination of the suggested control system. The findings demonstrate the effective performance of the suggested control mechanism against nonlinear effects and uncertainties.