Fast finite time adaptive neural network control for a class of uncertain nonlinear systems subject to unmodeled dynamics

• Published in: Information sciences Vol. 565; pp. 306 - 325
• Main Authors: Zhang, Yan, Wang, Fang, Yan, Feng
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.07.2021
• Subjects: Adaptive neural network control; Backstepping; Fast finite time control; Unmodeled dynamics; Backstepping; Unmodeled dynamics; Fast finite time control; Adaptive neural network control
• ISSN: 0020-0255; 1872-6291
• DOI: 10.1016/j.ins.2021.02.048

In this paper, a fast finite time adaptive control issue is discussed for a class of uncertain nonlinear systems. The systems considered involve unmodeled dynamics as well as dynamical disturbances. First, neural networks (NNs) are introduced to deal with the difficulties caused by unknown nonlinear uncertainties, and dynamical signal functions are utilized to handle unmodeled dynamics and dynamical disturbances. Second, based on adaptive backstepping technique and a fast finite-time stability criterion, a fast finite time adaptive neural network control scheme for a class of uncertain nonlinear systems subject to unmodeled dynamics is proposed for the first time. The proposed control scheme can not only ensure that all closed-loop signals are bounded, but also has the robustness to unmodeled dynamics and dynamical disturbances. The main innovations of this work lie in the ingenious design of parameter adaptive laws and controller, and the development of fast finite-time adaptive control algorithm from a new point of view. Finally, the feasibility of the proposed control algorithm is elaborated by two simulation examples.