Robust Adaptive Synchronization Control for a Class of Perturbed and Delayed Neural Networks

This paper is concerned with the asymptotic synchronization problem of a general neural network using the robust adaptive control technique. It is considered a class of modified Cohen–Grossberg neural networks which is supposed to undergo unknown perturbations caused by state-independent nonlinearities and bounded mixed time-varying delays on neuron amplification and activation functions. An adaptive compensation control strategy is proposed to ensure the elimination of the perturbed and delayed effects by means of adaptive estimations of unknown controller parameters. Through Lyapunov stability theory, it is shown that the proposed adaptive compensation controllers can guarantee the asymptotic synchronization of neural networks without knowing the knowledge of bounds of nonlinearities and delays. A numerical example is provided to illustrate the effectiveness of the developed techniques.