Dynamic Event-Triggered Adaptive Neural Output Feedback Control for MSVs Using Composite Learning

• Published in: IEEE transactions on intelligent transportation systems Vol. 24; no. 1; pp. 787 - 800
• Main Authors: Zhu, Guibing, Ma, Yong, Li, Zhixiong, Malekian, Reza, Sotelo, M.
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Actuators; Adaptive control; adaptive neural network; Adaptive systems; Artificial neural networks; Classification; classification reconstruction; Closed loops; Control theory; disturbance observer; Disturbance observers; event-triggered control; Feedback control; Learning; Marine surface vehicles; Neural networks; Output feedback; Reconstruction; State observers; Surface vehicles; Technological innovation; Uncertainty; Vehicle dynamics; Wear
• ISSN: 1524-9050; 1558-0016; 1558-0016
• DOI: 10.1109/TITS.2022.3217152

This paper investigates the control issue of marine surface vehicles (MSVs) subject to internal and external uncertainties without velocity information. Utilizing the specific advantages of adaptive neural network and disturbance observer, a classification reconstruction idea is developed. Based on this idea, a novel adaptive neural-based state observer with disturbance observer is proposed to recover the unmeasurable velocity. Under the vector-backstepping design framework, the classification reconstruction idea and adaptive neural-based state observer are used to resolve the control design issue for MSVs. To improve the control performance, the serial-parallel estimation model is introduced to obtain a prediction error, and then a composite learning law is designed by embedding the prediction error and estimate of lumped disturbance. To reduce the mechanical wear of actuator, a dynamic event triggering protocol is established between the control law and actuator. Finally, a new dynamic event-triggered composite learning adaptive neural output feedback control solution is developed. Employing the Lyapunov stability theory, it is strictly proved that all signals in the closed-loop control system of MSVs are bounded. Simulation and comparison results validate the effectiveness of control solution.