Dr Licio Romao
Parks Road, Oxford OX1 3QD
Feedback is a central concept in the study of dynamical systems, as it enables robustness against modelling and measurement errors. Such a concept has led to tremendeous advancements in several industries, e.g., aeronautical and aerospace, chemecial, biology, to name a few. Standard feedback design techniques, however, rely on mathematical models to design optimal controllers, and thus cannot be applied to applications for which the underlying model is too complex or may not be obtained using physical principles.
My research combines techniques from formal verification, control theory, applied mathematics, and machine learning to enable the design of safer and more reliable feedback systems. It focuses on addressing theoretical and practical aspects of the decision-making process under uncertainty. My research interests include:
- Dynamic programming and reinforcement learning
- Abstraction of dynamical systems
- Formal verification
- Stochastic systems
- Stochastic optimisation
- Feedback control systems
Licio Romao is a posdoctoral researcher at the Department of Computer Science, University of Oxford, since April 2021. He is a member of Oxcav where he works with Prof. Alessandro Abate at the intersection of machine learning, verification, and control. He obtained his PhD in June 2021 from the Department of Engineering Science, Univeristy of Oxford, under the supervision of Prof. Antonis Papachristodoulou and Prof. Kostas Margellos. He obtained his MSc in Electrical Engineering from the University of Campinas (UNICAMP), Brazil, in 2017, under the supervision of Prof. Pedro Luis Dias Peres, and BSc in Electrical Engineering from the Federal University of Campina Grande (UFCG), Brazil, in 2014. His PhD thesis has been awarded the Institute of Engineering Technology’s (IET) Control and Automation Dissertation Prize 2021, which recognizes annualy the best UK thesis in the broad field of automation and control.
His reserach interests include verification of discrete time stochastic systems by means of abstraction techniques, reinforcement learning, theoretical foundations of optimization algoritms, especially in the presence of uncertainty, and the design of feedback control loops.