Research
Main research interest: complementary AI approaches. The use of AI versus classical methods is sometimes a contentious area. There are classes of problems that can be solved already using classical methods and are well-understood and modeled. I would like to see more AI used to complement existing classical or provable controls approaches. This was the focus of two internships at the Navy Research Lab in Washington DC in 2015 and 2016, where among other projects I researched ways to extend linear temporal logic control to large numbers of agents to make it more scalable. Additionally, this approach is a core focus of my dissertation, which focuses on using AI for high-level traffic management, which fits on top of a robotic system that uses optimal path-finding algorithms.
Broader research interests: heterogeneous systems. I've looked at ways of incorporating methods from agent modeling into large heterogeneous multiagent systems, and extending the approaches of agent modeling and stereotyping to systems with complex dynamics and large numbers of agents. Recent research has focused on investigating the potential effectiveness of modeling strategies in the Unmanned Aerial Systems (UAS) conflict-avoidance domain. This was the focus of two internships at NASA Ames Research Center during the summers of 2012 and 2013.
Past research in Design Engineering has focused on using multiagent systems to identify creativity in past designs. This was the topic of my master's thesis, which was accepted in June 2013 by the Oregon State University Mechanical Engineering Graduate School. This work was accepted as a full paper in the ASME Design, Computing and Cognition conference. I also have worked toward developing better reward strategies to promote coordination in complex domains such as multiagent network routing.