My research develops foundations for the control of cyber-physical systems (CPS) -- and applies them to healthcare.
CPS represent a new era of safety-critical embedded systems that feature tight coupling between communication and
computation used to control complex, uncertain, and potentially adverse physical plants. In critical infrastructures (e.g.
healthcare, energy, transportation, manufacturing) existing robust engineering paradigms do not address the
combination of challenges imposed by closed-loop CPS: strict safety and performance requirements, constrained sensing and
actuation capabilities, unreliable communication and computation platforms, and potentially malicious feedback information.
In medical CPS, these challenges also include unidentifiable non-linear time-varying physiological processes
that result in significant model uncertainty -- an unfavorable environment for closed-loop control.
My research focuses on control systems engineering for CPS, encompassing control system design, analysis, and implementation.
The long-term goal enables high-assurance control of CPS through new algorithm designs, which consider implementation
constraints, such that the resulting real-world system facilitates analysis. Achieving this goal -- especially in
healthcare -- requires control system designs that conform to the platform resources while also providing predictable
performance in spite of complex cyber-physical interactions. Quantifying the trade-offs between performance, safety,
reliability, and security in the control designs requires new analysis tools and techniques. Validating control system
design assumptions requires real-world implementations that provide evidentiary support and enable identifying unforeseen
challenges and future research.