Temporal-Intermittent-Spatio-Sampling Control for Stochastic Switched RDNNs With Impulsive Effects

Abstract

This brief mainly investigates the stabilization problem for a class of state-dependent switched reaction-diffusion neural networks (RDNNs) with stochastic disturbance and impulsive effects. First, a hybrid control strategy of time domain intermittent-spatial domain point sampling is proposed, which can greatly reduce the control cost. Then, by using the constructed multi-partition Lyapunov function and the convex combination method, two different stabilization criteria are derived for the target system with stabilizing impulse and destabilizing impulse, which are shown in the form of easy-to-calculate LMIs. Moreover, the multi-gain phenomenon is effectively avoided through the single decision-multivariable technique, which facilitates the operation of the actual system. Note that our results are more practical than those established only on pure stabilizing impulse and destabilizing impulse. Particularly, one of the results shows that even if the switched RDNN with destabilizing impulse is unstable, it can be stabilized by the hybrid intermittent sampling control strategy proposed in this brief. Finally, numerical examples show the rationality and validity of the new results.