Adaptive NN control for a class of discrete-time nonlinear systems based on input-output model

A direct adaptive control structure, inverse controller based on input-output model, is studied for a class of SISO discrete-time nonlinear systems in this paper. The developed controller and adaptation algorithm are very simple to implement. The closed-loop system has been proven to be semi-globally uniformly ultimately bounded (SGUUB) if the design parameters are suitably chosen under certain requirement. An arbitrarily small tracking error can be achieved if the size of neural networks is chosen large enough. Simulation studies verify the effectiveness of the newly designed schemes and the theoretical discussion.