Robust estimator design for periodic neural networks with polytopic uncertain weight matrices and randomly occurred sensor nonlinearities

• Published in: IET control theory & applications Vol. 12; no. 9; pp. 1299 - 1305
• Main Authors: Rao, Hong-Xia, Xu, Yong, Zhang, Bin, Yao, Deyin
• Format: Journal Article
• Language: English
• Published: The Institution of Engineering and Technology 12.06.2018
• Subjects: control system synthesis; filtering theory; Lyapunov function; Lyapunov methods; matrix algebra; neurocontrollers; periodic neural networks; polytopic uncertain connection weight matrices; polytopic uncertain weight matrices; polytopic vertices; randomly occurred sensor nonlinearities; Research Article; robust control; robust estimator design; sufficient conditions; uncertain systems; uncertain systems; filtering theory; periodic neural networks; polytopic vertices; control system synthesis; polytopic uncertain connection weight matrices; robust estimator design; matrix algebra; sufficient conditions; neurocontrollers; randomly occurred sensor nonlinearities; robust control; polytopic uncertain weight matrices; Lyapunov methods; Lyapunov function
• ISSN: 1751-8644; 1751-8652; 1751-8652
• DOI: 10.1049/iet-cta.2017.1163

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_\infty $H∞ performance for the augmented system are established, and the corresponding gains of the estimator are designed. Finally, an illustrative numerical example is given.