Observer-based finite-time bounded analysis for switched inertial recurrent neural networks under the PDT switching law

摘要

In this note, the observer-based finite-time boundedness analysis issue for the switched inertial recurrent neural networks (SIRNNs) is investigated deeply. The switching law, persistent dwell-time, with more generality and universality is employed. The first target is to develop a switched estimation system (SES) to obtain the states from the output of the researched open-loop SIRNNs. Thereafter, based on the before-mentioned SES, the resulting switched estimation error system (SEES) without the extrinsic disturbance, along with the closed-loop SIRNNs under state feedback controller are constructed. Furthermore, the sufficient conditions that the exponential stability for the SEES and the finite-time boundedness for the closed-loop SIRNNs are established simultaneously. The relevant estimator and controller gains are deduced by a straightforward decoupling manner. Ultimately, the feasibility of the method proposed is clarified and illustrated via a numerical example. (C) 2019 Elsevier B.V. All rights reserved.