Vibration control of suspension system using proposed neural networks

This paper investigates a new robust model based neural controller for active suspension system's vibrations via feedback control approach. The proposed model reference adaptive control system consists of a neural controller, a robust feedback controller, a third-order linear reference model and dynamics of active suspension system. The simulation examples with various standard input signals are included to demonstrate the effectiveness of the proposed control method and show significant improvement over the existing PID controller method. The robustness of the proposed neural controller is also analyzed with white noise disturbances on the suspension system. It is shown that the control system is robustly stable for all road disturbances. Finally, this kind of control approach could be employed in real time vehicle applications.