Electrohydraulic active suspension neural-based control system

This paper presents an artificial neural networks (ANN) and proportional-integral-derivative (PID) controller design for electrohydraulic active vehicle suspension model, considering nonlinear dynamics of the actuator. First, the PID controller is used to estimate the suspension control force required for the system stabilization. An ANN configuration is proposed to deal with the nonlinear actuator dynamics. Two ANN configuration is used by mapping the actuator model and training it, for the force-tracking problem by a voltage control signal estimation. Simulations and the obtained results for the active suspension show that the control scheme proposed can deal with the actuator nonlinearity and improve the suspension performance.