A non-ellipsoidal reachable set estimation for uncertain neural networks with time-varying delay

The problem of bounding the reachable set for a class of neural networks in the presence of both polytopic uncertainties and time-varying delay is addressed in this paper. The maximal Lyapunov functional, combined with the Razumikhin methodology, is utilized to give a non-ellipsoidal description of the reachable set. An optimization algorithm is proposed in order to make the description as accurate as possible. This method shows great advantages compared to the existing ones. Finally, two examples are given to show the merits of our proposed method.