These Thought-Controlled Robotic Arms Are Beating Paralysis and Amputation

In 2012, University of Pittsburgh researchers released a video of Jan Scheuermann feeding herself a bite of chocolate. This, of course, wouldn’t be noteworthy but for one thing: Scheuermann is paralyzed from the neck down. She fed herself that chocolate using a brain implant and thought-controlled robotic arm—and got a taste of freedom once unthinkable.

Scheuermann’s spinocerebellar degeneration left her unable to move her limbs over a decade ago. She leapt at the chance to take part in the University of Pittsburgh study investigating brain-computer interfaces. The study’s researchers are developing a system that reads and decodes brain activity, translating it into physical action in a robotic arm and hand.

Mindblowing as that early video was, it showed the brain-computer interface at a fairly rudimentary stage. Scheuermann could move the robotic arm’s shoulder, elbow, and wrist, but the hand was only able to open and close.

In a paper detailing the last two years of the study, the researchers describe how the system has advanced since 2012. Instead of its original seven degrees of freedom, it now has ten degrees of freedom including a range of new hand positions.

The researchers weren’t sure their software could handle the added complexity. They started Scheuermann out with a virtual robotic arm on a computer display. The software learned to recognize patterns of brain activity and associate them with arm and hand positions. All the while, Scheuermann got more adept at controlling the arm.

Now, instead of one object of uniform shape and size, Scheuermann can manipulate items of varying shapes and sizes. The system isn’t ready for primetime—but gives a tantalizing glimpse of the future. Though she knew the day was coming when her role in the research would end, Scheuermann said she “bawled” for 15 minutes after they removed the implants.

Robotic prosthetics are making notable gains too. In a recent Johns Hopkins video, double amputee, Les Baugh, controls two robotic arms with his thoughts. Baugh’s arms work on a slightly different principle than Scheuermann’s. Instead of relying on a brain implant to interpret commands, the arms use signals from surgically reassigned nerve endings.

The technique, called targeted muscle reinnervation, works by reassigning nerves associated with various arm and hand movements to new muscles. The nerves still fire—even though they aren’t connected to an arm or hand anymore. When Baugh thinks of moving his arm or hand, his muscles contract, and the system learns to associate various patterns of contraction with the appropriate commands for the robotic arm and hand.

Like Scheuermann, Baugh started training in a virtual environment.

The researchers involved were impressed at how quickly he mastered the arms. After just ten days, he was performing fairly complicated movements, like for example, transferring an empty cup between shelves, a task requiring eight different motions. This is extraordinary, in part, because Baugh lost his arms 40 years ago.

Further, while we’ve covered similar robotic prosthetics before, none we’ve seen worked with a double shoulder-level amputee. Baugh can perform complex movements and rapidly switch between the two arms.

“It’s like the early days of the Internet,” said Michael McLoughlin, Johns Hopkins principal investor. “I think the next five to 10 years are going to bring phenomenal advancement.”

An exciting prospect. But one perk to being part of the research now? Next steps include giving Baugh a pair of the systems to take home with him to see how they might aid his everyday life.

It’s little things most people take for granted that Baugh is most looking forward to trying. “Maybe for once I’ll be able to put change in the pop machine and get pop out of it,” he said.

Image Credit: University of Pittsburgh/YouTube

Jason Dorrier
Jason Dorrier
Jason is editorial director of Singularity Hub. He researched and wrote about finance and economics before moving on to science and technology. He's curious about pretty much everything, but especially loves learning about and sharing big ideas and advances in artificial intelligence, computing, robotics, biotech, neuroscience, and space.
Don't miss a trend
Get Hub delivered to your inbox