Developing autonomous flight vehicles for NASA

James Madison University computer science alumnus, Benjamin Kelley (’13), is working to improve decision-making machines as part of NASA’s Autonomy Incubator (AI) housed at the Langley Research Center in Hampton, Virginia. Kelley has officially been the team's software architect and developer since June of 2015. The AI program consists of “multi-disciplinary scientists, engineers, and student interns” working to develop autonomous flight vehicles, according to an article published on NASA’s website.

Kelley first became involved through his position at Analytical Mechanics Associates, a lead contracting company at NASA’s Langley Research Center. “I was placed on the Autonomy Incubator task force because of my experience with C++, architecture design, and the [Data Distribution Service] messaging system,” Kelley said.  

According to Kelley, the group focused on two goals. “We were tasked with bringing in new hires and developing their autonomy research expertise to help with projects all over the center; and we set out to further the state of the art of autonomous systems.” Research focused on innovations to solve common issues faced by autonomous system developers.

Kelley’s studies at JMU helped prepare him for his role in the Autonomy Incubator as a software architect and developer. “The coursework I completed in the Computer Science (CS) Program challenged me and helped me learn not only how to code, but more importantly how to problem solve,” Kelly said. “I was routinely just outside my comfort area, which forced me to learn and grow as a computer scientist.”

As a software developer, Kelley credits his expertise to CS Professor, Christopher Fox. “His course on software design was one of the most impactful courses I took at JMU,” said Kelley. “Fox taught us that having a well-designed system makes it easier to maintain, easier to understand, and easier to extend and improve.”

In addition to Fox, professors Sharon Simmons and Florian Buchholz, guided Kelley’s studies. “I worked on a research project with Simmons for the Security and Software Engineering Research Center,” said Kelley. Kelley learned not only about classic software development, “but how to tackle problems that few people in the larger computer science community are working on.”

Buchholz advised Kelley as he wrote his thesis during his graduate studies. “He was a great source of inspiration and guidance,” said Kelley, “while working on what was arguably the hardest project during my time at JMU.” Buchholz taught Kelley the importance of attention to detail, which has aided Kelley in his current work. “In an autonomous system, one small mistake or a slight inefficiency in an algorithm can snowball into a much larger error in the overall system.”

Reflecting on his experience with AI, Kelley stressed the importance of open-mindedness. “Even the most mundane software development tasks can teach you something and help improve your skillset.” He pointed to his own career as an example of his philosophy. “I never set out with the ultimate goal of working with quadcopters and autonomy,” said Kelley, “but my previous projects as a CS student, gave me the skillset which led to this amazing opportunity.”