Banca de QUALIFICAÇÃO: VITOR VALE PEREIRA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : VITOR VALE PEREIRA
DATE: 02/10/2020
TIME: 09:00
LOCAL: Google Meet: meet.google.com/znb-kwds-tio
TITLE:

Intelligent control of a ball-beam system using artificial neural networks


KEY WORDS:

Non-Linear Control, Neural Networks, Machine Learning, Feedback Linearization, Ball-Beam, TinyML, Microcontroller


PAGES: 40
BIG AREA: Engenharias
AREA: Engenharia Elétrica
SUMMARY:

This work describes the development of a non-linear control strategy for a ball-beam system. Interesting aspects of the this system is its open loop instability, strong nonlinearity and the fact that it is an under-acted system. The adopted strategy is based on a non-linear control technique combined with a machine learning technique based on artificial neural networks, which have been famously used in systems that need to emulate the human inference process. Thus configuring an intelligent control, as it is able to predict the uncertainties of the system based on past experiences and to adapt to changes in the system and the environment. The TinyML paradigm is used for the implementation of this technique, which proposes the integration of mechanisms based on machine learning powered by microcontrollers. With the rapid growth of the Internet of Things, objects with embedded microcontrollers are becoming more present in our lives, so providing them with cognitive skills opens up an infinite number of opportunities for the development of a broad collective intelligence network.


BANKING MEMBERS:
Presidente - 1445637 - WALLACE MOREIRA BESSA
Interno - 1328152 - CARLOS EDUARDO TRABUCO DOREA
Interno - 1242315 - PABLO JAVIER ALSINA
Notícia cadastrada em: 16/09/2020 16:21
SIGAA | Superintendência de Tecnologia da Informação - (84) 3342 2210 | Copyright © 2006-2024 - UFRN - sigaa10-producao.info.ufrn.br.sigaa10-producao