Data Driven Learning and Control seminar series is organized by the Information and Decision Science Lab at Cornell University and aims to explore the latest advancements and interdisciplinary approaches to data-driven learning and control systems.
Watch on YouTube Live
Supervisory Reference Governor Schemes for Safe Control of Uncertain Systems and for Model Predictive Controllers
Reference governors are add-on predictive safety supervision algorithms. They monitor and modify, when necessary, the commands passed to the nominal system to ensure that state and control constraints are satisfied and safety is preserved. After introducing reference governors, the talk will touch upon several recent developments in reference governor theory and applications to constrained control. This includes the learning reference governor, which safely learns by experimenting with an uncertain system subject to state and control constraints. Additionally, the talk will discuss the use of reference governors to enforce constraints for controllers generated through data-driven construction of Control Lyapunov Functions. Developments in the use of reference governors as add-on supervisory schemes for Model Predictive Control (MPC), aimed at reducing computational time and enlarging the constrained closed-loop region of attraction, will also be described.
Bio: Ilya Kolmanovsky earned his Ph.D. in aerospace engineering (flight dynamics and control) from the University of Michigan in 1995, alongside an M.A. in mathematics completed the same year. He also holds an M.S. in aerospace engineering from Michigan (1993). His doctoral research focused on motion planning and feedback control for nonholonomic dynamic systems, with applications to attitude control of underactuated multibody spacecraft.
Kolmanovsky joined the University of Michigan’s Department of Aerospace Engineering in January 2010 where he has been appointed as a full professor (with tenure) since September 2013 and as a Pierre T. Kabamba Collegiate Professor of Aerospace Engineering since September 2023.
His current research aims at advancing control theory for systems with state and control constraints, including Model Predictive Control and Reference Governors. He also focuses on the modeling, dynamics, and control for advanced spacecraft, aircraft, automotive vehicles and engines and propulsion systems.
Before returning to the academia, Kolmanovsky spent nearly 15 years at Ford Research and Advanced Engineering in Dearborn, MI, progressing from postdoctoral researcher to technical leader in powertrain control. His work at Ford centered on control of advanced internal combustion engines and powertrain systems to improve transient response and drivability, increase fuel and energy efficiency, and reduce emissions.