Imagen de Portada

Spatial Trajectory Tracking of Underactuated Autonomous Underwater Vehicles by Model–Data-Driven Learning Adaptive Robust Control

Guardado en:
Publicado en:Journal of Marine Science and Engineering vol. 13, no. 6 (2025), p. 1151-1175
Autor principal: Guo Linyuan
Otros Autores: Zhou, Ran, Guo Qingchang, Ma Liran, Hu Chuxiong, Luo Jianbin
Publicado:
MDPI AG
Materias:
Kinematics
Robust control
Mathematical models
Feedback
Adaptation
Tracking
Line of sight
Guidance (motion)
Adaptive control
Parameter uncertainty
Tracking errors
Autonomous underwater vehicles
Compensation
Disturbances
Learning
Tracking control
Neural networks
Controllers
Feedforward control
Design
Underwater vehicles
Parameter estimation
Environmental
Acceso en línea:Citation/Abstract
Full Text + Graphics
Full Text - PDF
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Resumen:This paper aims to solve the spatial trajectory tracking control problem of underactuated autonomous underwater vehicles (AUVs) in the presence of system parameter uncertainties and complex external disturbances. To accomplish this goal, a model–data-driven learning adaptive robust control (LARC) strategy is introduced for AUVs. Firstly, a serial iterative learning control (ILC) approach is introduced as feedforward compensation, and then the corresponding trajectory tracking error dynamics model, the Feedforward Compensation–Line of Sight (FFC-LOS) guidance law, and the feedforward compensation-based kinematics controller are designed. Secondly, the dynamics controller is designed for AUVs, which consists of a linear feedback term, a nonlinear robust feedback term, an adjustable model compensation term, and a fast dynamic compensation term. In this control framework, the robust control and fast dynamic compensation parts are utilized to deal with nonlinear uncertainties and disturbances, the projection-type adaptive control part solves the influence caused by the uncertainty of system parameters, and the serial ILC part that is a data-driven learning method can further improve the trajectory tracking accuracy for repetitive tasks. Finally, comparative simulations under different scenarios and different types of disturbances are performed to verify the effectiveness of the proposed control strategy for AUVs.
ISSN:2077-1312
DOI:10.3390/jmse13061151
Fuente:Engineering Database