DoD using AI, machine learning to speed up improvements to uncrewed vehicles: official
NewsApril 23, 2024
SAN DIEGO, California. The Department of Defense is using artificial intelligence and machine learning to speed up software improvements to uncrewed vehicles across the services from a matter of months to a few days, a DoD official said at the XPONENTIAL 2024 exhibition on Tuesday.
While working on the Lionfish uncrewed underwater vehicle (UUV) program, operators found they needed to make updates to the software, which involved pulling the software out of the UUV and then mailing it to engineers who would work on it and then sending it back six months later, said Doug Beck, director of the Defense Innovation Unit.
"One of the things that the team discovered as they were working [with the Lionfish UUV] was that there are critical updates needed to the software from training missions to refining the capability to improving search and avoidance," Beck said. "That's like a six-month process. Obviously we don't have six months."
So the team implemented AI and machine learning tools to speed up the software improvement process, which allowed them to cut the process from six months to just one to two weeks -- and in a recent exercise with the Australians, that time was reduced to six days.
"That's a dramatic change there, and it's based on artificial intelligence," Beck said.
Now, the DoD is looking to stretch that capability across the services so that other programs can benefit.
"We're working with the chief digital and artificial intelligence officer on standardizing that," Beck said. "We're looking at how we take this capability to the other domains as well."
