Design of adaptive controller with a neural networks compensator for of the nonlinear dynamic systems

This work presented adaptive controller with auxiliary controller using neural network for unknown nonlinear systems. The proposed controller is composed of an approximate controller (PID controller) and a neural network auxiliary controller. Our method is different from those using supervised learning algorithms, such as the back-propagation (BP) algorithm, that needs training information in each step. Simulation results show the output tracking error between the plant and the reference model can asymptotically converge to zero in the presence of disturbances and approximation errors. The contributions of this paper are the new approach to constructing neural network architecture and its training.