Neuromorphic adaptive generalized predictive controller (NAGPC) for the tracking problem

A neuromorphic adaptive generalized predictive control (NAGPC) is presented. This controller is based on the use of neural networks for the adaptive tuning of the output horizon in the generalised predictive control (GPC). The choice of the output horizon in the GPC is critical for the tracking problem of variable reference signals. With the controller proposed this problem is overcome and a satisfactory tracking response can be achieved in the system. Some simulations has been carried out to validate the efficiency of this method.