PhD Opening in Physics-Informed Machine Learning, Control, and Optimization at University of Central Florida University of Central Florida in United States

Truong Xuan Nghiem at the University of Central Florida is recruiting one PhD student for the Intelligent Cyber-Physical Systems (iCPS) Lab in Physics-Informed Machine Learning, Control, and Optimization.

The project focuses on building composable, physics-informed learning methods for complex cyber-physical systems, including differentiable and modular models of large-scale heterogeneous systems, principled ML methods that integrate physical constraints, data, and domain knowledge, and learning-enabled control and optimization with safety, reliability, and real-world impact.

Possible application areas include autonomous cyber-physical systems such as smart energy systems, robotics, and networked infrastructure. The post emphasizes both theory and implementation: developing new methods, building reusable software frameworks, and translating research ideas into practical tools.

Preferred background: MS in engineering, computer science, applied math, or a related STEM field; exceptional BS students may be considered. Research experience in machine learning, control, optimization, robotics, or autonomous CPS is preferred, especially if demonstrated through publications or substantial projects. Strong programming and computational skills are valued, particularly in Python, Julia, differentiable programming, optimization, or scientific computing.

Eligibility note: Domestic US students or international students already in the US are especially encouraged to reach out.

How to apply: Submit an application through the UCF graduate portal and email the professor with the subject line "PhD EE application" along with a brief intro, relevant background, CV, publications, and supporting documents. The post says only shortlisted applicants will be contacted.

Research context: The announcement is aligned with recent work on physics-informed machine learning for modeling and control, differentiable causal block diagrams, and safe physics-informed machine learning for dynamics and control.