H ∞ State Estimation for Switched Inertial Neural Networks With Time-Varying Delays: A Persistent Dwell-Time Scheme

The $\mathcal {H}_{\infty }$ state estimation issue for switched delayed inertial neural networks is addressed in this work. A more universal switching law, persistent dwell-time (DT) switching law, is considered here rather than average DT one of which switching frequency among subsystems is strictly limited, or DT one. Concurrently, time delays are inevitable when transmitting information, then taking the time-varying delays into account makes the constructed systems conform well with the actual situations. The main goal in the work is devoted to designing a state estimator to ensure that the state estimation error system is globally uniformly exponentially stable and satisfies a prescribed $\mathcal {H}_{\infty }$ noise attenuation level. A mixed time/mode-dependent Lyapunov–Krasovskii functional matched with the foregoing switching law is introduced. Through utilizing some reasonable inequalities and common matrix operations, some sufficient criteria which guarantee the aforesaid stability and the solvability of the addressed issue are presented. Finally, an illustrative example is provided to present the potentiality and validity of the developed results.