My statement of research interests, experience, and future plan is available in PDF format.
Below is a short summary of my research interests and approach.
My research advances the scientific foundations and practical applications of Cyber-Physical Systems (CPS) – engineered systems with seamless integration of computational algorithms and physical components.
CPS are driving innovation in important sectors such as energy, transportation, manufacturing, healthcare, and agriculture, to achieve the vision of Smart and Connected Communities that will transform our lives.
However, the challenges for CPS are significant due to the complex and deep interdependencies among engineered systems and the natural world.
At the same time, with the recent advances in computer science, control, and computing technologies, this area is poised for explosive growth with far-reaching opportunities.
My work integrates control, optimization, machine learning, and computation to address fundamental CPS challenges across various domains, with a focus on Cyber-Physical Energy Systems (CPES).
Data-driven CPS Modeling and Control: CPS require new models that can capture the complex relationships between the cyber and physical components and can evolve with the systems. My work harnesses the availability of data and the power of machine learning to develop effective and flexible data-driven modeling approaches as well as to close the loop of data-driven CPS modeling and control for controlling CPS efficiently in the presence of uncertainty.
Managing Uncertainty in CPS: Handling uncertainty is critical for CPS because uncertainty is ubiquitous, especially in large-scale systems, and presents a significant challenge in modeling, controlling, and optimizing CPS. I develop optimal control systems and new scheduling algorithms to exploit the synergy of interconnected buildings for managing uncertainty and volatility in power grids.
CPS Design and Verification: Greater autonomy of CPS demands greater assurances of performance, safety and reliability, placing a high premium on systematic design and verification methods. I optimize the efficiency and ensure the safety of CPS through optimal control-computation co-design and patented practical verification methods.
CPS Tools: Tools and platforms are instrumental in the technology and engineering of CPS, putting the CPS science into applications. I have developed open-source software for co-simulations of CPS to facilitate research and development in smart energy systems, among which are MLE+ and openBuildNet. MLE+ won the best demo award at BuildSys – a premier conference on smart built environments.
My research approach uses theoretical insights and techniques from diverse areas as the basis to develop new theories, algorithms, and systems.
A key priority is to find and interact with the right domain experts outside of my core disciplines as it ensures high impact of my research.
I find such interdisciplinary work particularly exciting and educational.