Matthew Gombolay is the Catherine M. and James E. Allchin Early-Career Assistant Professor in the School of Interactive Computing. Gombolay’s research interests span robotics, artificial intelligence (AI), machine learning (ML), human factors engineering and human-robot interaction (HRI). His passion is exploring the entire pipeline, from deriving analytical equations, developing algorithms, and fielding robots for human-robot teaming experiments.
In his research, Gombolay takes the perspective that there are “super users” or individuals with decades of experience whose decision-making abilities often outperform the best efforts of computer scientists. “In aerospace final assembly, every minute of downtime on the factory line can translate to hundreds of thousands of dollars in lost revenue opportunity,” he explains. “In the factory, there are people recognized as legends for their ability to coordinate factory logistics and keep production lines moving smoothly.”Gombolay and his colleagues develop novel machine-learning techniques to capture and replicate the actions of these super users, which can then be leveraged within high-performance computing to scale beyond the users’ expertise.
As a licensed pilot, Gombolay has a vested interested in developing algorithms that allow for human-in-the-loop dynamic route planning and collision avoidance in a congested air space. “Distributed coordination of activities across aircraft is key to the successful implementation of a dense urban and regional air mobility airspace” he says. “Having personally experienced what it is like to fly in congested air space, we need to make sure that as urban and regional air mobility operations increase, we develop coordinated, intelligent, distributed systems that support pilot and air traffic controller decision-making.”
Learn more about Gombolay’s research here.