Purdue team places fourth in NSWC Crane ‘AI for Small Unit Maneuvers’ competition
The team earned prize money against more established and well-funded defense, academic and industrial competitors.
Twenty participants were invited to present their Phase I White Paper concepts to the Artificial Intelligence for Small Unit Maneuver (AISUM) Prize Challenge judging panel. Seven were selected to participate in Phase II – Virtual Vignettes for the Artificial Intelligence for Small Unit Maneuver (AISUM) Prize Challenge. Phase II participants were asked to develop specific algorithms that were used to compete in virtual scenarios and were evaluated for their algorithms used within a government- provided virtual map.
Tyler Fedrizzi, a senior in Electrical Engineering and co-founder of Codex, led the team.
“Having the opportunity to work together as a team to tackle this very difficult challenge was an incredible way to spend the summer,” Fedrizzi says. “The entire team worked extremely hard and overcame so many obstacles, which furthered our own knowledge while also helping all of us to improve our communication skills, increase our technical proficiency, and engage in the diverse research ecosystem with NWSC Crane. The winnings from this challenge will allow our team to continue developing our work, with the ultimate goal of producing a system that keeps military members safe and secure.”
Other members of the team included senior Aerospace Engineering students Arpit Amin and Saketh Nibhanupudi and Cole Kniffen, a junior in Electrical Engineering. Shreyas Sundaram, Marie Gordon Associate Professor in the Elmore Family School of Electrical and Computer Engineering, and Shaoshuai Mou, associate professor of aeronautics and astronautics, who are co-directors of the Center for Innovation in Control, Optimization, and Networks (ICON), helped advise the team.
“This was an intense competition, requiring the teams to solve very difficult problems involving artificial intelligence, computer vision, robotic path planning, and autonomous systems in a compressed time frame, Sundaram says. “The fact that Tyler and his team were able to do so well, especially against large teams from industry, is extremely impressive, and speaks volumes about the caliber of the engineering students at Purdue.”
Artificial Intelligence for Small Unit Maneuver (AISUM) combines Naval Expeditionary Warfare and Special Operations Forces (SOF) tactical maneuver elements with Robotic Autonomous Systems (RAS) to create a low-risk, human-machine maneuver element that gains, maintains and extends access in complex, contested and congested areas, providing a decisive advantage against adversaries.
Over the past 20 years, SOF have become proficient at utilizing tactical elements and overhead remotely piloted airborne ISR systems. However, the threat landscape has changed, with sophisticated adversaries that can leverage Electronic Warfare (EW) tools and complex, congested environments especially in urban settings. This necessitates new technologies that can enable SOF to re-establish dominance against such threats. This competition, and the solution provided by the Purdue team, moves us in that direction. Specifically, the Purdue team has taken the first steps to show how a fully autonomous Unmanned Aerial System can navigate a complex urban environment, and provide helpful and timely intelligence to humans, enabling them to safely and effectively clear threats.