Improved Integral Sliding Mode Control for Adaptive Synchronization of Delayed Recurrent Neural Networks

Abstract

This paper is devoting to addressing the problem of adaptive synchronization for recurrent neural networks with time-varying delays, in which the unknown parameters are taken into account. A new and simple integral sliding mode control (SMC) strategy is developed motivated by the previous idea of the existing literature. Based on drive-response concept, the synchronization error system is firstly derived such that the corresponding synchronization error signal can be applied to construct a simple integral sliding mode manifold. Then, an appropriate sliding mode controller with adaptive laws is designed such that the reachability of the predefined manifold is ensured and thereafter the synchronization problem between the response system and a drive system with unknown parameters is addressed. Finally, two numerical examples are given to substantiate the effectiveness of the presented control strategy.