Perlis Leads Team that Wins $1M DARPA Award for AI Research

Three researchers in the University of Maryland Institute for Advanced Computer Studies (UMIACS) have won an award of nearly $1 million to develop artificial intelligence that will improve the decision-making capabilities and contextual awareness of robots working cooperatively.

The funding—$980,000 over 18 months—comes from the Defense Advanced Research Projects Agency (DARPA), a branch of the Department of Defense that develops emerging technologies for the military. The funding is intended to lay the foundational research for an interoperable AI tool that can work independently from humans, learn from its mistakes, and complete tasks that may arise over the course of a mission by coming up with creative solutions.

Don Perlis (in photo), a professor of computer science with an appointment in UMIACS, is the lead investigator on the project. He is joined by Justin Brody and Tim Clausner, who are visiting research scientists in UMIACS.

The three UMIACS members bring their own areas of expertise to the multidisciplinary project. Perlis’s work involves AI and robots, Brody’s specialty is in machine learning and model theory, and Clausner’s research is centered on human cognition and human-computer interaction.

The goal of the Context Reasoning for Autonomous Teaming (CREATE) program is to teach autonomous systems to react to changing contexts and make decisions as a group without human direction. For instance, if a communication network goes down, the team of robots would still be able to undertake a mission on its own.

–Story by Maria Herd