Set-Based Asynchronous State Estimation for Networked Switched Neural Networks
研究了一类离散时间切换神经网络在事件触发机制下的区间估计问题，采用集合成员估计方法处理异步性，提高了估计精度和适用性，并给出了切换信号与观测器的协同设计。
This study addresses the zonotopic interval estimation for a class of discrete-time switched neural networks (NNs) under an event-triggered mechanism. Unlike the existing studies, it adopts zonotopic set-membership estimation and considers general asynchronism, thereby enhancing the accuracy and applicability of the estimation results. First, by constructing appropriate observer-mode-dependent Lyapunov functions, certain sufficient conditions are established to guarantee the stability and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">${\ell }_{\infty }$</tex-math> </inline-formula> performance of the augmented error system. Based on these conditions, we present a co-design approach for the switching signals and event-triggered nonlinear observer. Furthermore, a time-varying state zonotope is established, and the corresponding estimated bounds are derived. Finally, the efficacy of the proposed interval estimation approach is validated through two illustrative examples.