Reachable set estimation of neural networks with stochastic sampled-data
Te Yang, Zhen Wang, Guoliang Chen, Jianwei Xia · 2022 13th Asian Control Conference (ASCC) · 2022
In this paper, the reachable set estimation (RSE) and stochastic sampled-data (SSD) controller design of neural networks with external disturbances and time-varying delay are investigated. Firstly, the loop-based Lyapunov functional is constructed via introducing two stochastic variables whose occurrence probability satisfies Bernoulli distribution, and the sufficient condition that all states are bounded under zero initial condition is obtained. Secondly, the RSE is considered in the SSD controller design, and the goal is that the ellipsoid contains the reachable set of the closed-loop system. Finally, the effectiveness of these methods is verified by a numerical example.