UNM Postdoctoral Fellow: Rosalyn Devonport

Most days for Rosalyn Devonport are spent in her office working advanced and abstract math problems, throw in some computer programming, and if she's having a really stimulating day, coffee with a friend is one of her first stopping points - you'd be surprised by how much advanced math is done over a freshly brewed cup of coffee.

While Devonport's day-to-day seems routine, the 30-year-old is a University of New Mexico postdoctoral fellow working to discover what it takes for a spacecraft to obtain "orbital homeostasis."

"It's the ability for a satellite to remain in orbit in spite of its inherently chaotic nature, and to protect itself from any of the things in space that would remove it from that orbit, of which there are more than you might think," Devonport said.

Ultimately, Devonport's research is an interdisciplinary branch of engineering mathematics called control theory, which she applies to the study of spacecraft control systems for satellites stationed in a distant part of earth's orbit called cislunar space.

"I can describe the basics of control theory with something we are all familiar with, the human body," Devonport said. "The temperature outside can be 15 degrees or 110 degrees; you can be walking, running, or lying down but your body will remain a reliable 98.6 degrees. Just by being yourself, you embody the property of 'being robustly self-regulating against changes to the environment.' The scientific name for this property is called homeostasis and control theory, at its core, is the study of how to endow machines with the property of homeostasis."

Devonport received her doctoral degree in electrical engineering from the University of California at Berkeley. She has co-authored nearly two dozen papers, half of which Devonport is the primary author. She has also co-authored the book, Interval Reachability Analysis. During her time at Berkeley, Devonport was introduced to UNM Professor Meeko Oishi, and with similar fields of interest, Devonport found herself moving to the Southwest.

"Dr. Oishi happened to be looking to recruit postdocs, and I happened to be looking to become one," Devonport said. "It's as simple as that, really. I'm here because a top researcher in my field wanted me here, and we were both interested in doing the same kind of work."

It's a familiar position most of our 127 postdocs find themselves in when looking for a fellowship. Devonport's postdoctoral scholarship resides within the UNM School of Engineering. However, control theory is closely related to mathematics. Devonport said rather than studying addition, subtraction, multiplication, and division, she spends time with advanced mathematics which is the study of patterns, particularly patterns that occur in nature.

According to Devonport, within her area of study, there are basically three problems with making a cislunar satellite self-regulating.

1. Dynamics are uncertain: There are subtle gravitational effects of the sun and moon that are difficult for any candidate artificial cerebellum to compensate for.
2. Researchers never know when a strong gust of solar radiation pressure will push a satellite around.
3. These satellites are very difficult to observe, since they are so far away, so researchers are dealing with a scenario with very limited information.

"If we can't account for all three at once, we cannot stabilize the satellite," Devonport said. "The stakes could hardly be higher, because any failure risks the destruction of an incredibly valuable satellite. I am a domain expert in the first two problems -- controlling uncertain systems and controlling systems with a random component -- and the goal of my postdoctoral research is to find a way to combine them. Control theorists have developed amazing ways to account for these problems over many decades, but we tend to look at them in isolation."

Devonport said, within her field, she's recently taken on a new reputation as a good scientific communicator.

"I'm known as the lightning talk queen," Devonport said. "One of my special skills is finding a narrative thread by which any research topic can be described. By finding the story, it becomes much easier to communicate complicated concepts to folks that are unfamiliar with them."

A skill she put into practice during this interview…

"I could have taken the easy way out and just said that I do research on spacecraft," Devonport said. "It would be correct to say so, and it's fascinating to talk about. But it wouldn't be true to the spirit of my work, because I work with something even more fascinating than space, I work with mathematics."