New results on adaptive fixed-time control for convex-delayed neural networks

Shengqin Jiang; Yukun Song; Weili Zeng; Haokui Zhang; Shuiming Cai; Xiaobo Lu

doi:10.1016/j.isatra.2022.08.027

New results on adaptive fixed-time control for convex-delayed neural networks

ISA Trans. 2023 Mar:134:134-143. doi: 10.1016/j.isatra.2022.08.027. Epub 2022 Sep 1.

Authors

Shengqin Jiang¹, Yukun Song², Weili Zeng³, Haokui Zhang⁴, Shuiming Cai⁵, Xiaobo Lu⁶

Affiliations

¹ School of Computer Science, Nanjing University of Information Science and Technology, Nanjing, 210044, China; Engineering Research Center of Digital Forensics, Ministry of Education, Nanjing University of Information Science and Technology, China. Electronic address: jiangshengmeng@126.com.
² School of Computer Science, Nanjing University of Information Science and Technology, Nanjing, 210044, China; Engineering Research Center of Digital Forensics, Ministry of Education, Nanjing University of Information Science and Technology, China.
³ College of Civil Aviation, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, China.
⁴ Intellifusion, Shenzhen 518057, China.
⁵ Faculty of Science, Jiangsu University, Zhenjiang 212013, China.
⁶ School of Automatic, Southeast University, Nanjing 210096, China.

PMID: 36109253
DOI: 10.1016/j.isatra.2022.08.027

Abstract

This paper studies the adaptive fixed-time synchronization issue for convex-delayed neural networks. First, the convex delay is introduced to address the state delay of neural networks in order to reflect the impacts of multiple delay components such as input transition time and switching communication. Then, a new fixed-time control method is presented to adaptively determine multi-control gains with a unified update law. Afterward, some sufficient criteria are figured out by using Lyapunov stability theorem to ensure that the delayed neural networks are fixed-timely stable. Finally, simulated examples are adopted to validate our theoretical results.

Keywords: Adaptive control; Convex delay; Fixed time; Neural networks.