Department Seminar with Javad M. Velni: Supervised Learning for Control-oriented Modeling and Anomaly Detection in Complex Cyber-physical Systems: Challenges and Opportunities

When

March 5, 2020    
10:00 am - 11:30 am

Where

3043 ECpE Building Addition
Coover Hall, Ames, Iowa, 50011

Event Type

Speaker: Javad M. Velni, Associate Professor of Electrical Engineering at the University of Georgia

Title: Supervised Learning for Control-oriented Modeling and Anomaly Detection in Complex Cyber-physical Systems: Challenges and Opportunities

Abstract: Cyber-physical systems (CPS) link cyberspace with the physical space through a network of interconnected elements, such as sensors, actuators, communication and computational engines (control and data analytics). While (first principles) model-based approaches are commonly used in CPS design and control, they become inadequate in capturing the complexity as systems become larger and extremely dynamic, and subsystems involved become collaborative. The adaptive and dynamic nature of emerging CPSs makes data-driven approaches highly desirable. This talk aims at describing our recent work on two problems of paramount importance to the control and resilience of CPSs. First is the development of new analytical, data-driven tools for the model identification of complex CPSs by leveraging supervised statistical machine learning. Second is the development of tools and algorithms for detecting anomalies in large-scale networks, where a fusion of graph-theoretic approaches with graphical learning methods are presented to reduce the computational complexities involved in solving underlying estimation problems and hence for detecting anomalies. Two prominent examples of CPSs examined in this talk are intelligent automotive systems and smart grids. Application of the developed methods for modeling and anomaly detection for these two CPSs will be examined. We will conclude the talk by providing a roadmap for the application of data-driven tools (graph theory and machine/deep learning) in various theoretical and applications settings.

Bio: Dr. Javad Mohammadpour Velni received BS and MS degrees in electrical engineering from Sharif University of Technology and University of Tehran, Iran, respectively, and PhD degree in mechanical engineering from University of Houston, TX. He is currently an associate professor of electrical engineering at the University of Georgia (UGA), where he joined in 2012. Prior to that, he was with the University of Michigan, where he worked in the naval architecture & marine engineering department from Oct. 2011 to Jul. 2012. He was also a Research Assistant Professor of mechanical engineering at University of Houston from Oct. 2008 to Sep. 2011 and a Research Associate at the same institution from Jan. 2008 to Sep. 2008. He has published over 160 articles in international journals and conference proceedings, served in the editorial boards of ASME and IEEE conferences on control systems and edited two books on control of large-scale systems (published in 2010) and LPV systems modeling, control and applications (published in 2012). His current research interests are in secure control of cyber–physical systems (and in particular, smart grids and transportation systems), coverage control of heterogeneous multi-agent systems, and learning-based control of complex distributed systems.

Seminar Host: Degang Chen

Loading...