Output feedback NN control of NARMA systems using discrete Nussbaum gain

In this paper, output feedback adaptive neural network (NN) control is investigated for a class of discrete-time NARMA (nonlinear-autoregressive-moving-average) system. To solve the noncausal problem in control design, the system is transformed by future outputs prediction. The difficulty of nonaffine appearance of the control input is overcome by exploiting of implicit function theorem. The lack of a priori knowledge on the control directions is solved by using discrete Nussbaum gain. The high-order-neural-network (HONN) is employed to approximate the unknown ideal control. The closed-loop system achieves semi-global-uniformly-ultimately-boundedness (SGUUB) stability and the output tracking error is made within a small neighborhood around zero by suitably choosing the design parameters. Simulation results are presented to demonstrate the effectiveness of the proposed control approach.