Abstract

Recent technological advances have made devices for actuation, sensing, computation, and communication increasingly portable, inexpensive, and prevalent in societal engineering network systems. Examples include the power grid -and its emergent energy markets-, intelligent transportation systems, connected autonomous vehicles, and robotic networks. The emerging use of purely data-driven mechanisms to control and optimize in real time these complex network systems has led to the awareness of the pitfalls of model-free decision making without stability and robustness guarantees. This limitation is further exacerbated by the complex interactions that emerge between the continuous-time dynamics and the discrete-time dynamics of the closed-loop system, which difficult the development of rigorous stability, convergence, and robustness certificates via control theoretic tools. To address these challenges, in this talk I will present some of our recent advances in the context of feedback control with data-assisted feedback loops, with a focus on nonsmooth and hybrid control approaches. The proposed controllers are suitable for the solution of model-free optimization problems in complex dynamical systems subject to topological constraints, high-performance requirements, and safety demands. The algorithms exploit non-Lipschitz and hybrid (continuous and discrete) dynamics to overcome fundamental limitations of standard smooth adaptive algorithms, achieving accelerated model-free control without sacrificing critical stability and robustness guarantees. Extensions to model-free time-varying decision making in game theoretic settings will also be discussed in the context of coordinated network games. Applications to transportation systems, robotic networks, and power grids will be presented to illustrate the main theoretical results.  

Bio

Jorge I. Poveda is an Assistant Professor in the Department of Electrical, Computer, and Energy Engineering at the University of Colorado, Boulder. He received the M.Sc. and Ph.D. degrees in Electrical and Computer Engineering from UC Santa Barbara in 2016 and 2018, respectively. Before joining CU Boulder in 2019, he was a Postdoctoral Fellow at Harvard University, and a summer research intern at the Mitsubishi Electric Research Laboratories, Cambridge, MA. Dr. Poveda has received the CCDC Outstanding Scholar Fellowship and the Best Ph.D. Dissertation award from UC Santa Barbara, the Best Student Paper Finalist award at the IEEE Conference on Decision and Control (as student in 2017 and as co-author in 2021), the Research Initiation award (CRII) by the NSF (2020), the Young Investigator Program (YIP) award by the Air Force Office of Scientific Research (2022), and the RIO Faculty Fellowship at CU Boulder (2021-2022). His research interests include hybrid and nonlinear control theory, adaptive and model-free optimization, and applications in cyber-physical systems.  

Hosts

Philip E. Paré, philpare@purdue.edu