Feature
Austin Whitney didn’t want to graduate from college in a wheelchair. So he and the student engineers at U.C. Berkeley’s “Kaz Lab” built a machine that allowed him to stand up and walk across the commencement stage

Long Walk: Strapped into an exoskeleton that he helped build, Austin Whitney walks across the stage at his U.C. Berkeley graduation. Below, from left: Michael McKinley, Jason Reid and Wayne Tung of the Robotics and Human Engineering Lab stand behind him.  Dmitri Alexander

General Electric unveiled the world’s first powered exoskeleton—Hardiman, a 1,500-pound, marginally functional beast—in 1965, but it was too bulky and difficult to control to be practical. Exoskeletons now often weigh less than 100 pounds, but they work fundamentally the same way. The user is strapped into a metal framework with articulated limbs and joints that he directs with body movements or electrical stimulation. Motors supply the superhuman power.

Whitney’s exo is named “Austin” in his honor, and its robotic heritage extends back to BLEEX, an exoskeleton that Kazerooni began developing in 2000 with grant money from the Pentagon’s Defense Advanced Research Projects Agency (Darpa). In 2005, Kazerooni co-founded the company Berkeley Bionics to commercialize the devices pioneered in his university lab.

With the goal of developing an exoskeleton that costs close to what a powered wheelchair does, the students were forced to adopt a minimalist approach. Instead of updating earlier models by adding features, they had to strip them out. “Kaz told us, ‘Find a way to use fewer motors and still make this thing walk,’” Pillai says. She and the team employed two motors rather than the four or six motors of previous models. Instead of custom-designing every part, the team members bought some components from Sports Authority, including snowboard bindings and soccer shin guards to secure Whitney’s feet and legs, and shoulder straps from a backpack to hold the exo’s motors and batteries to his back. Reid, who was in charge of the Austin’s computer control systems, had to work with a system that had very few actuators powering joint movements or sensors reporting leg positions to the computer brain, which itself is a $60 microprocessor rather than a $500 one. “It’s been a huge adjustment to get performance out of the machine,” he says.

Seven minutes into the test, Whitney has walked only about 20 feet. The team had completely rebuilt the knees just a few days prior, a major design change equivalent to replacing the wings of the space shuttle a week before takeoff. The good news is that they’re holding firm. But now the machine’s torso framework is out of alignment. “It just doesn’t feel right,” Whitney says. “It’s pushing really hard on my lower back, and that’s causing me pain.” He takes another step, then stops, panting, head hung down over his shoulders and a bead of sweat dripping from his nose. “I can’t make it any further,” he says. “Get the chair."

McKinley and Tung fix the back-angle problem by 2 a.m. But new problems keep cropping up. On Thursday, after working flawlessly for thousands of previous steps, a bolt in the hip joint wriggles loose and locks the legs into a completely straight position. Austin has to be lifted, like a corpse frozen in rigor mortis, back to his chair. At Friday evening’s dress rehearsal, less than 24 hours before the commencement ceremony, Whitney’s black graduation gown gets pinned beneath the pads that hold him in the exo. When he tries to stand up, the tautly stretched fabric breaks one of the plates in the back. This time the Kaz Lab members are up until 3 a.m. machining a replacement.

“It’s like a game show,” McKinley says. “Like someone has written these riddles for us and we have x-amount of time to solve them.” When Whitney stands at graduation, though, the time to solve problems will have expired. If you comb YouTube for exoskeleton videos, what you’ll see are tightly edited demonstrations in clinical settings where every variable, such as the walking surface, can be controlled. Whitney, in contrast, is going to perform live and outdoors. “This will be a very transparent demonstration,” Reid says, “for better or worse.”

single page

7 Comments

Analysis indicates that a strategy is being missed. That the interface can be as simple as a small body modification performed without a surgeon. We could all be virtual Cyborgs in less than a year if the missed strategy works. However, conductive silicone, and or combined with conductive hydrogels will make the full dream happen. Blastocyst in pigs to grow organs and limbs is the other technique that is coming soon.

You can not invest enough money in this type of tech period

One key to affordable robotics will be micro and nano electrical systems that can be cheaply self assembled. As they say, right now, all these pieces are custom crafted with machine tooling and probably custom prototyping equipment.

As far as the specific design, if it is really as complicated a process to get a reliable gait then they need to start thinking about modifying the mode of locomotion to something easier, instead of trying to simulate walking with cheaper equipment. For example, they already have devices like gyroscopically balanced unicycles that allow a person to move around basically by shifting upperbody weight. I would bet that such devices could be modified so that they could even go up (or down) shallow stair steps.

Its a great achievement. Rather a shame that it came after millions of dollars of research for a military machine instead of the reverse. Think of what we could achieve if we had our priorities straight.

For cheap robotics innovations they might consult with toy manufacturers who are always working to get the most bang for the buck. They would also know how to mass produce a machine economically.

Amazing!!!
I am VERY happy for the young man whom science was able to assist his ambulatory abilities. Moore's law states that in 18 months the abilities of this assistive technology will increase by at least double.
As an emergency medical technician I've been at the scenes of accidents similar to the one he described having caused his condition. I am hopeful that a day shall dawn that these injuries are a thing of the past. Barring that technological advancement this is a step in the right direction to restore persons to mobility.

@weird science
Moore's law is only true for the most basic component of computers. For example, even though processors have constantly gotten faster, has your word processor gotten better at the same rate?

This is more like a concept demo more than anything. Lower body exoskeletons have many problems, and will eventually be solved, but it will take a while.

For example:
This was walking straight on a flat surface, the world is not straight and flat. The exoskeleton must be able to either take clues from the user to position the legs or navigate itself.
Exoskeletons will first have their uses in assisting humans that still have use of their legs, but not enough. This could be because we want the person to be stronger than the normal human, or because the person has weakened legs for some reason.

Imagine someone who has been unable to use their legs for a while, then regains use of them. Without the muscle tone and bone mass the person must undergo a lot of physical therapy to be able to walk again. Instead we could give the person an exoskeleton that allows them to perform everyday tasks and slowly decrease the amount of assistance the exoskeleton provides.



June 2013: American Energy Independence

Five amazing, clean technologies that will set us free, in this month's energy-focused issue. Also: how to build a better bomb detector, the robotic toys that are raising your children, a human catapult, the world's smallest arcade, and much more.


Online Content Director: Suzanne LaBarre | Email
Senior Editor: Paul Adams | Email
Associate Editor: Dan Nosowitz | Email
Assistant Editor: Colin Lecher | Email
Assistant Editor: Rose Pastore | Email

Contributing Writers:
Rebecca Boyle | Email
Kelsey D. Atherton | Email
Francie Diep | Email
Shaunacy Ferro | Email

circ-top-header.gif
circ-cover.gif