Intelligent semi-active vehicle suspension systems using neural networks

A methodology is proposed for designing intelligent vehicle suspension systems which implement the magneto-rheological damper as an active element. The adaptive control law is constructed based on the neural network methodology and Taylor series approximation. The controller commands the current of the magneto-rheological damper and controls directly the damper force. The neural network is trained with respect to a developed road disturbance scenario and its parameters are obtained using a global numerical optimisation technique. The proposed control law has a novel structure capable of sensing the interactions between the variables and thus can adjust the feedback gains with respect to the existing conditions. It takes into consideration the actuator's dynamics and avoids limit cycling which is caused by the hysteretic behaviour of the MR dampers. The performance of the intelligent system is evaluated by means of simulations in MATLAB for quarter and half car models.