University of Arizona researchers are leading a three-year, $4.5 million Pentagon contract to study artificial intelligence for hypersonic aircraft navigation. Hypersonics is becoming an increasingly critical field in global defense.
The project's principal investigator, Systems and Industrial Engineering professor Roberto Furfaro, is joined by AME professors Samy Missoum and Jesse Little and AME assistant professors Alex Craig and Kyle Hanquist.
“We built these facilities for these types of projects,” said Turbulence and Flow Control Laboratory director Little, in an interview with the Arizona Daily Star. “They are for large laboratory projects, multiple users, involving government, industry and people involved in different disciplines.”
According to the Daily Star, the researchers will use data gathered from computer simulations and wind-tunnel tests about how vehicles behave in hypersonic flow to create a simulated environment for training the adaptive “brain” of the system. To train hypersonic systems to navigate and react to extremely complex, high-speed situations on their own, the UA team is using a type of machine learning called meta-reinforcement learning.
The research project is sponsored by Pentagon’s Joint Hypersonic Transition Office through the University Consortium for Applied Hypersonics.
“With meta learning, we can train it not only on one scenario, but on many scenarios,” Furfaro said. “The system is able to learn over a distribution environment, and every time it converges faster to the next one. By enabling this continuous learning, we are basically able to have a system that continually adapts.”