Eventos Anais de eventos
DINAME2019
DINAME2019
Feedback Linearization and Supervisioned Neural Networks for the Depth Control of a ROV
Submission Author:
Diago Cesar Xavier de Freitas Barros , RN , Brazil
Co-Authors:
Diago Cesar Xavier de Freitas Barros, Gabriel da Silva Lima, Wallace Moreira Bessa
Presenter: Diago Cesar Xavier de Freitas Barros
doi://10.26678/ABCM.DINAME2019.DIN2019-0212
Abstract
A ROV (Remotely Operated underwater Vehicle) system needs a precise estimation of its position in order to avoid damage and imprecise movements in certain missions. In this paper, the intelligent control of the path trajectory of a ROV susceptible to external forces (water dynamics, currents, animals, etc.) is proposed based on feedback lin-earization and use of an error estimator with neural network. This work proposes an approach to the matter using a series of different neural networks, under supervised learning, for the unknown dynamics, to evaluate the performance of each network case, both in time and error prediction between the ideal trajectory and the real due to disturbances.
Keywords
ROV, Intelligent control, feedback linearization, neutral networks
