Pular para o conteúdo

27/03/2025 – Defesa de Tese de Doutorado – Eduardo Scheffer Saraiva

Defesa Pública da Tese de Doutorado do Mestre em Engenharia Elétrica Eduardo Scheffer Saraiva

Data da defesa: 27/03/2025 – quinta-feira
Horário: 13h30min

Área de Concentração: Automação e Controle
Linha de Pesquisa: Controle

A defesa será realizada através de videoconferência pelo link:  https://meet.google.com/kcu-jodc-kgc

Banca Examinadora:
Prof. Dr. Daniel Ferreira Coutinho (Relator) – DAS/PPGEAS – UFSC
Profa. Dra. Grace Silva Deaecto – FEM – UNICAMP
Prof. Dr. João Manoel Gomes da Silva Jr. – PPGEE – UFRGS
Presidente da Banca/Orientador: Prof. Dr. Aurélio Tergolina Salton – PPGEE – UFRGS
Coorientador: Prof. Dr. Jeferson Vieira Flores – PPGEE – UFRGS

 Title: “INTERNAL MODEL CONTROL OF UNCERTAIN EULER-LAGRANGE SYSTEMS”

Abstract:
“This thesis presents a systematic approach to the modeling and control of Euler-Lagrange systems, with an emphasis on periodic signal tracking. The control problem is framed as a convex optimization task using a differential-algebraic representation (DAR) and Linear Matrix Inequalities (LMI), which are central to the methods developed in this thesis. These approaches provide formal guarantees of bounded error trajectories and form the theoretical foundation of the proposed control framework. The study begins with the development of an optimal controller for fully known systems and progresses to a robust design that incorporates an internal model capable of handling uncertainties while ensuring practical stability. Extensions to the control framework are introduced to address actuator limitations, incorporating saturation and anti-windup mechanisms. These enhancements ensure the controller’s applicability in practical scenarios where actuators operate within physical constraints, while maintaining stability and performance. The proposed algorithms are validated on the benchmark system of a KUKA LWR-IV+ robot, where they are compared with traditional controllers such as sliding mode and computed torque. A comprehensive analysis of the results highlights the advantages and limitations of the approach for different controllers in real-world robotic systems.

Keywords: DAR-Descriptor, Robust Output Regulation, Robotic Manipulator, LMIs, Convex Optimization, Practical Stability.”