Abstract: Cyber-physical systems (CPS) like autonomous vehicles and smart grids are becoming ubiquitous in our society. As CPS grow in complexity, machine learning (ML) is being increasingly integrated to enhance their performance and autonomy. While ML has shown great potential, it alone cannot guarantee performance, safety, and reliability in many, sometimes mission-critical, CPS. Our research aims to develop intelligent CPS by seamlessly integrating ML with physics, control, optimization, and computing.

 

In this talk, I will first discuss the need for synergistically blending ML and physics in CPS. I will then provide an overview of our recent work on data-driven modeling, active learning, learning-based control, and adaptive sampling to improve CPS efficiency. I will also highlight our current and future research to integrate physics knowledge into ML for physics-informed learning and control. Integrating first principles from physics provides many advantages like interpretability, data efficiency, accuracy, and safety guarantees compared to data-driven ML alone. Finally, I will demonstrate applicability of our approaches in building HVAC systems and robotics.

Short Bio: Dr. Truong X. Nghiem is an Assistant Professor in the School of Informatics, Computing, and Cyber Systems at Northern Arizona University. He received his Ph.D. in Electrical and Systems Engineering from the University of Pennsylvania. Dr. Nghiem's research focuses on developing integrated frameworks leveraging control, optimization, machine learning, and advanced computing to tackle challenges in cyber-physical systems. His expertise lies in data-driven methods for modeling, control, and optimization of autonomous systems. He received the National Science Foundation (NSF) CAREER Award in 2023 for his research on physics-informed machine learning for cyber-physical systems. He is also a recipient of the NSF ERI Award (2022) and the Best Paper Award at the ACM/IEEE International Conference on Cyber-Physical Systems (2018). Dr. Nghiem is a Senior Member of the IEEE and a member of the ACM.