Air Force Research Lab engineers attack manufacturing problems in the new Collaborative Automation for Manufacturing Systems, or CAMS, Laboratory.

At the push of a computer button, a Boston Dynamics robot comes to life, unlatching from a charging dock and striding confidently across a small room at the Air Force Research Laboratory on four padded feet, approaching visitors.

A few minutes later, the robot opens a lab door with robotic jaws, using one of its legs to keep the door open as its walks on its remaining three legs into the lab.

Sean Donegan, AFRL’s digital manufacturing research team lead at Wright-Patterson Air Force Base, assures the visitors the robot is harmless.

Still, the sight of a robot moving like a dog swiftly and unswervingly across a tiled floor is unnerving.

Said Donegan, “It’s an impressive piece of hardware.”

AFRL’s Collaborative Automation for Manufacturing Systems (“CAMS”) lab on Wright-Patterson Air Force Base’s Area B is where a small cadre of engineers and researchers pursue digital solutions to manufacturing problems.

The room itself is relatively small ― maybe 450 square feet. But some of the problems attacked here loom large: How can the Air Force better equip operators and technicians? How can the service more quickly produce tools and weapons airmen and maintainers need on the battlefield? Where can working prototypes leap from the laptop to the runway?

Can robots like this make impossible jobs less impossible?

The CAMS lab is the AFRL Digital Manufacturing Research Team’s first internal laboratory, meant to answer those questions and others.

“This lab is sort of our on-site research lab for all of the work that we do,” Donegan said.

Here, researchers aren’t doing basic research for everybody. They’re performing applied research with specific objectives and end-users in mind.

Part of what sets this lab apart are the tools. Here, autonomous manufacturing, including “collaborative robotics” and “extended reality” hardware and headsets can be used to approach problems from new angles.

Virtual and augmented reality headsets help operators plot how to train a robot to paint a plane or reattach a panel.

Not all of the lab’s work is performed here. CAMS works with university researchers. It worked with Ohio State University on a new way to forge bulk metal components.

Often, forged parts are manipulated into shape with a large press. One of the most common types of hot metal forging is closed-die forging with a mold. But the Department of Defense can face significant lead times in obtaining those parts — 12 to 48 months sometimes. That’s a big challenge for the Air Force (and other services) that need to field systems rapidly.

AFRL researchers began to examine the potential of “open-die” forging, a process that uses hammers and dies to manipulate metal into shape, not all that different from traditional blacksmithing, Donegan said.

But this is blacksmithing with a difference.

“Instead of having a human do it, we have a closed-loop robotics system do this,” he said. “It’s sort of a robotic blacksmith.”

The process resulted in a prototype, one that worked. And the potential of collaborative robotics — from Air Force “Collaborative Combat Aircraft” AI pilots to assistants on the ground — is gaining attention.

The robot cost AFRL about $200,000. That’s not inexpensive, but it’s a number Donegan says requires context.

“In the grand scheme of things, it’s both a lot and not a lot,” he said. To obtain an especially sensitive electron scanning microscope, for example, could cost $1 million or maybe more.

“In that context, it’s actually a pretty good investment, I think.”