Minisymposium on Vistas in Control

date: Friday, October 20th, 2023

time: 9:30am – 12:10pm

location: Engineering Science Building 2001, UC Santa Barbara

Organized by: Francesco Bullo, Mechanical Engineering, UCSB

Schedule (35 minutes per talk, 5 minutes for Q&A)

Speakers

Muruhan Rathinam

Professor, Department of Mathematics and Statistics, University of Maryland Baltimore County, Maryland

Abstract: Flocking is the dynamic behavior where a set of autonomous agents moving in a Euclidean space accelerate themselves in a manner that keeps them together. The well-known Cucker-Smale (CS) model assumes that each agent “j” exerts an acceleration on every other agent “i” which is proportional to the relative velocity of “j” with respect to “i”. It is also assumed that the proportionality coefficient is symmetric in “i” and “j” and is a fixed function of the distance between “i” and “j”. Flocking was defined mathematically by the condition that the diameter in the position space of the set of agents remains bounded in time and that the diameter in the velocity space approaches zero asymptotically in time. Conditions were provided for the flocking of the CS model. Motsch and Tadmor (MT) introduced a generalization which removed the symmetric influence assumption and new tools were introduced for the analysis of flocking. In this work, we provide a generalization of the MT model which can account for partial masking and orientation bias. Moreover, we show that a model akin to MT cannot exhibit the milling behavior where agents rotate around a common axis. To better account for reality, we add steering terms into our model which accounts for the individual whims of agents in response to externalities. We provide sufficient conditions for flocking under steering and also provide an asymptotic analysis of the situation where steering is much slower than flocking.

Biosketch: Muruhan Rathinam is a Professor of Mathematics at the University of Maryland, Baltimore County (UMBC). His research interests lie in the modeling and analysis of deterministic as well as stochastic dynamical systems, and also in the development and analysis of Monte Carlo methods for Stochastic Reaction Networks. Muruhan Rathinam obtained his PhD in Applied Mathematics from Caltech (1997) under the supervision of Prof. Richard Murray and held a postdoc position at Caltech (1997-1999). Prior to joining UMBC in 2002, he was a postdoc at University of California, Santa Barbara in the group of Prof. Linda Petzold (1999 - 2002).

Andrew D. Lewis

Abstract: I revisit the problem on which I wrote my first paper with Richard, namely that of comparing two ways of writing equations of motion for mechanical systems with nonholonomic constraints.

Biosketch: Andrew received his PhD under the sun of Professor Richard Murray in 1995. After two years as a postdoctoral fellow at Warwick University, Andrew joined the Department of Mathematics and Statistics at Queen’s University, Kingston in 1998. He coauthored a book with Professor Francesco Bullo in 2004.

Robert T. M’Closkey

Professor, Mechanical and Aerospace Engineering, Samueli School of Engineering and Applied Science, University of California, Los Angeles

Abstract: The development of a contactless platform for studying the dynamics of micro-scale resonators is reported. Electrostatic suspension of the platform —a silicon disk— is achieved using a transduction subsystem consisting of transformer-coupled electrode pairs. The electrodes exert electrostatic forces on the disk and also measure differential capacitances related to the disk position. Generically, the system has three unstable modes, however, certain electrode-disk geometries are not strongly stabilizable, which presents further challenges. The seminar will discuss the modeling, controller synthesis based on coprime factor representations of the plant, development of high fidelity models from closed-loop test data, and, most importantly for the study of micro-scale systems, measurement uncertainty due to transducer noise.

Biosketch: Robert T. M’Closkey received the Ph.D. degree in Mechanical Engineering from the California Institute of Technology and is currently a Professor in the Mechanical and Aerospace Engineering Department, University of California, Los Angeles. He develops micro-scale inertial sensors with a focus on Coriolis vibratory gyros (CVGs). He received the National Science Foundation CAREER Award for his research on MEMS inertial sensors and is a recipient of IFAC’s Mechatronics Journal paper prize.

Richard M. Murray

Abstract: The New Oxford American Dictionary defines “vista” as “a pleasing view, especially one seen through a long, narrow opening”. Using that as a starting point, this talk will attempt to describe how the “long, narrow opening” of control can be used to provide interesting perspectives on complex, engineered systems, across a broad range of applications and industries. I will attempt to simultaneous argue why control theory, as a discipline, is both outdated and critical for the future of technology. Along the way, I will talk about the use of concepts from control for wrangling complexity, the need for control theorists (and their siblings in dynamical systems) to embrace heuristics (or become irrelevant), and some of the open and closed challenges in the field. I’ll try to draw on my experience in academia, government (via advisory boards), and industry to create a pleasing view, as seen through a long, narrow opening. It should be fun, and hopefully not too offensive.

Biosketch: Richard M. Murray is a synthetic biologist and Thomas E. and Doris Everhart Professor of Control & Dynamical Systems and Bioengineering at Caltech, California. He was elected to the National Academy of Engineering in 2013 for “contributions in control theory and networked control systems with applications to aerospace engineering, robotics, and autonomy”. Murray is a co-author of several textbooks on feedback and control systems, and helped to develop the Python Control Systems Library to provide operations for use in feedback control systems.