Data-driven predictive control for linear parameter-varying systems

• Grant number: 25/12376-2
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: January 01, 2026
• End date: December 31, 2028
• Field of knowledge: Engineering - Electrical Engineering - Industrial Electronics, Electronic Systems and Controls
• Principal Investigator: Ricardo Coração de Leão Fontoura de Oliveira
• Grantee: Marcelo Sebastián Aravena Arellano
• Host Institution: Faculdade de Engenharia Elétrica e de Computação (FEEC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil

Abstract

Model Predictive Control (MPC) is a strategy that has been extensively studied and applied both in the scientific community and in industry, primarily due to its ability to handle operational constraints while optimizing the desired performance. With the remarkable advances in chip computational capabilities, in terms of both processing power and storage, data-driven control techniques, which do not require explicit knowledge of a model, have emerged as a new paradigm in the field of control. This PhD project proposes the development of data-driven predictive control techniques for linear parameter-varying (LPV) systems, aiming to provide conditions for the synthesis of stabilizing controllers that guarantee closed-loop performance. The formulation accounts for parametric uncertainties, parameter variation rates, actuator saturation, and the presence of disturbances. As a solution methodology, the project proposes the use of convex programming through linear matrix inequalities (LMIs), a versatile and well-established tool for robust analysis and synthesis of uncertain and time-varying parameter systems. (AU)