Robust Stability Criterion for Delayed Neural Networks with Discontinuous Activation Functions

摘要

The problem of global robust stability for a class of uncertain delayed neural networks with discontinuous activation functions has been discussed. The uncertainty is assumed to be of norm-bounded form. Based on Lyapunov–Krasovskii stability theory as well as Filippov theory, the conditions are expressed in terms of linear matrix inequality, which make them computationally efficient and flexible. An illustrative numerical example is also given to show the applicability and effectiveness of the proposed results.