A future fight against China will likely involve a network of bases spread out across small islands in the Pacific—a potential logistics nightmare. To prepare, the Air Force is working with Boeing on drones, AI, and augmented reality—empowered by 5G—that can make some basic maintenance tasks faster and less complicated.
At Joint Base Pearl Harbor-Hickam in Hawaii, members of the Air Force’s 15th Maintenance Group at last May tested a new way to perform maintenance tasks: using autonomous drones for routine aircraft inspections of aircraft, to look for things like corrosion, damaged rivets, etc.. and inform maintainers. That could drastically cut down on the time and complexity of inspections.
The drones could drastically cut down on the time and complexity of inspections—inspecting just the tail of a C-17 can take six hours, Scott Belanger, who works product support for Boeing Global Services, told reporters Tuesday.
“That includes getting everyone together, finding all the lifts, doing the safety briefs, putting the harness on, getting the helmet, realizing the lift doesn’t work. I mean, there’s just a whole process,” He said. And that timeline assumes the service doesn’t have to fly an inspector from another location.
Boeing’s early experiments in autonomous aircraft inspection show that an autonomous drone flying around the aircraft, finding trouble spots using image recognition artificial intelligence, and then pinging a human maintainer elsewhere to verify findings cuts the overall time down considerably: a routine inspection can be as short as 30 minutes.
Autonomous inspections would also give the Air Force a more up-to-date digital database of maintenance issues service-wide, which could help the force better predict where to station materials and maintainers in the future.
“You’ll be able to just pull up a tail number, click on, anywhere on that 3D model of that aircraft, and be able to see a history of images of that exact part you clicked on from anywhere in the world over the life of the aircraft,” explained Alli Locher, a Boeing senior product manager.
A drone working with humans is also more accurate than human inspectors alone, according to the first experiments. The Autonomous Aircraft Inspection system has an accuracy rate of about 70 percent; for humans alone, the accuracy is closer to 50 percent. And Boeing believes the system’s accuracy will improve over time, as maintenance experts use it more and tag more data.
A separate Boeing experimental effort to use 5G for maintenance tasks, dubbed ATOM, uses the Microsoft Hololens virtual reality headset to give maintenance crews visual cues to address problems and let other, more experienced technicians in other locations see what the technician is seeing.
So how much does virtual reality actually improve technician proficiency? Could you take an untrained airman off the runway, put her or him in a virtual reality headset, and turn them into a more skilled technician (rather than fly one in from elsewhere)?
Said Belanger, “I think we’re working with the customer right now to build that data.”
But that is the hope, he said.
“In the Air Force specifically there’s a three, seven, and nine level skill code. Could you take a three-level maintainer on a remote island in the Pacific and … using an ATOM headset, they can walk through some basic tasks that an engine [technician] would do? I firmly believe, as a 25-year maintainer, the answer is yes. That will work…We will go from zero capability—i.e. we’re having to fly will a whole nother C-17 out there with the right technician that may take 12 hours to get there—to: you can have an introductory-level maintenance conversation to make sure that when the follow-on augment crew does arrive, they’ve got the right tools. They’ve got the right equipment, the right skill set, and thus they can get to work immediately.”
Comments are closed.