This paper deals with the problem of course keeping control for marine vessel with limited communication band-width, unknown environment interference and model uncertainty. Firstly, for the Nomoto nonlinear model of marine vessel, the Backstepping method is used to deduce the course-keeping controller. By using the event-triggered condition, the rudder angle command is transmitted only when the condition is met, which can reduce the steering frequency of the rudder equipment. By taking advantage of radial basis neural networks (RBF-NN), unknown system terms and external disturbances can be compressed into an unknown parameter, and only one adaptive law needs to be updated online. Through Lyapunov analysis, it is proved that all tracking errors of the closed-loop system are globally consistent and finally bounded. Finally, the effectiveness of the algorithm is verified by numerical simulation.
Adaptive Neural Event-Based Course-Keeping Control of Marine Vessel
17.05.2024
360508 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
COURSE CONTROL SYSTEM FOR MARINE VESSEL, AND MARINE VESSEL
Europäisches Patentamt | 2021
|Neural Network for Course-Keeping Control
British Library Conference Proceedings | 1995
|Marine vessel with gyroscope-optimized station keeping
Europäisches Patentamt | 2023
|