Robust state estimation for periodic neural networks with uncertain weight matrices and randomly occurred sensor nonlinearities

This work addresses the problem of the estimator design for the periodic neural networks with polytopic uncertain connection weight matrices. The polytopic uncertainty is used to model the uncertain weight matrices. Bernoulli processes are employed to characterise the randomly occurred sensor nonlinearities, where the sensors are distributed in a large area. A Lyapunov function which depends both on the polytopic vertices and the period is constructed to improve the performance of the estimator. Sufficient conditions of the stochastic stability with H ∞ performance for the augmented system are established, and the corresponding gains of the estimator are designed. Finally, an illustrative numerical example is given.