The autonomous system cannot do that, though. If lethal force is required, a human will have to make that decision and enable the boats to fire on a target, Klunder emphasized.
"We have every intention of using these unmanned systems to engage a threat and destroy it if necessary," he said. "But there is always a human in the loop on the designation of a target and the destruction of a target. The power of what I'm holding here is the autonomous feature. They sense, avoid, go get and engage with a target. They work together."
For now, Brizzolara said his group is working to meet the Navy's goal of having these autonomous systems working in a year.
Their biggest challenges, he noted, are improving the system's perceptive abilities and the boats' ability to work together.
"There are still refinements that we need to do in terms of the major components of the system," said Brizzolara. "We need more reliable sensing so we see everything that's out there and more reliable route planning for multiple boats working together. If one boat needs to take itself out of the team to go investigate something else, the team needs to compensate for that one removed boat."
One of the biggest problems researchers face is making sure the boats are operating from the same information and that they are working together to decide what to focus on and how to move as a swarm.
"We've taken that first step, but there is still work we need to do," said Brizzolara. "We need to continue to develop the algorithms for the route planning and then we need to do simulation testing and then [go] out and test it on the water."
The Navy is looking to use the system around large naval vessels, merchant vessels, ports, harbors and oil rigs.
"We've taken a patrol craft usually manned with three or four sailors each and made it unmanned," said Klunder. "All those sailors normally on those crafts can be back on the ship and out of harms way."
Sign up for Computerworld eNewsletters.