This paper presents multi-layer feedforward neural network-based identification and approximate predictive controller (NNAPC) for atwo degree-of-freedom (DOF), quarter-car servohydraulic vehicle suspension system. The nonlineardynamics of the servo-hydraulic actuator is incorporated in the suspension model. A suspension travelcontroller is developed to improve the ride comfortand handling quality of the system. A SISO neuralnetwork (NN) model based on Nonlinear AutoRegressive with eXogenous input (NARX) is developed using input-output data sets obtained from mathematical model simulation. The NN model was trainedusing Levenberg-Marquardt algorithm. The NNAPCwas used to predict the future responses that are optimized by cost minimization. The proposed controller is compared with a constant-gain PID controller (based on Ziegler-Nichols tuning method) during suspension travel setpoint tracking in the presenceof deterministic road disturbance. Simulation resultsdemonstrate the superior performance of the NNAPCover the generic PID - controller in adapting to thedeterministic road disturbance.
展开▼
