Hasta La Gadget!

Aman Behal’s automated robotic arm functioned perfectly. Outfitted with sensors that could “see” objects, grasp them with enough force to hold but not crush them, and return them to the user, it easily outperformed the same arm under manual control on every quantitative measurement.

Except one. The arm’s users — patients with spinal cord injuries in an Orlando hospital — didn’t like it. It was too easy.

“Think about the Roomba,” Behal told Wired.com. “People like robots, and they like them to work automatically. But if you had to watch and supervise the Roomba while it worked, you’d get frustrated pretty quickly. Or bored.”

This wasn’t what Behal had expected. This was the new sensor’s system first time in the field; the user satisfaction survey was supposed to be one more data point, secondary to measuring the performance of the device itself. But it made his team rethink their entire project.

Behal’s arm is just one in a long line of robotic arms aimed at giving paraplegics and quadriplegics greater freedom and mobility. Recent advances have made robot arms far more sensitive, powerful, and realistic than ever before. In many cases the enhancements depend on software that allows the robot arms to take simple commands (or even signals from the user’s brain) and translate them into complex movements involving multiple motors without requiring their users to specify the exact movements of each servo. But in this study, Behal found that there’s such a thing as too much automation.

Behal, an Assistant Professor at the University of Central Florida, had initially used the arm in a 2006 study at the University of Pennsylvania funded by the National Science Foundation and the National Multiple Sclerosis Society. In addition to weakening physical control, MS often impairs attention and memory, and the complexity of the arm’s controls overwhelmed them. At that time, the arm’s sensors and AI were much more limited, and users were mostly frustrated by its complicated controls.

For these patients, according to Behal, something that might seem as simple as scratching their heads was a prolonged struggle. They needed something that took the guesswork of movement, rotation, and force out of the equation.

The quadriplegics at Orlando Health were the opposite. They were cognitively high-functioning, and some had experience with computers or video games. All had ample experience using assistive technology. Regardless of the extent of their disability or whether they were using a touchscreen, mouse, joystick, or voice controls, they preferred using the arm on manual. The more experience they had with tech, the happier they were.

It didn’t matter that the arm performed faster and more accurately when it was fully automated. Users were actually more forgiving of the arm when they were piloting it. If the arm made a mistake on automatic mode, they panned it. Harshly. (“You see a big vertical spike downward,” when that happened, Behal said.) On manual mode, the users learned how to operate it better — and how to explain their problems with the device to someone else.

To users accustomed to navigating the world in a wheelchair — and frequently having to explain how their chair worked to others — this made the arm both more familiar and more useful. It felt less like an alien presence, and more like a tool: a natural extension of the body and the will.

This feeling is essential for anyone’s satisfaction using technology, but particularly so for disabled users, according to John Bricout, Behal’s collaborator and the associate dean for Research and Community Outreach at the University of Texas at Arlington School of Social Work.

“If we’re too challenged, we get angry and frustrated. But if we aren’t challenged enough, we get bored,” said Bricout. He’s seen this repeatedly with both disabled and older adults.

In an interview with Wired.com, he expanded on this, drawing on psychologist Mihály Csíkszentmihályi’s theory of flow: “We stay engaged when our capabilities are matched by our challenges and our opportunities,” Bricout said. If that balance tilts too far to one direction, we get anxious; if it tilts to the other, we get bored. Match them, and we’re at our happiest, most creative, and most productive.

Behal and Bricout hadn’t anticipated, for example, that users operating the arm using the manual mode would begin to show increased physical functionality.

“There’s rehabilitation potential here,” Bricout said. Thinking through multiple steps to coordinate and improve physical actions “activated latent physical and cognitive resources… It makes you rethink what rehabilitation itself might mean.”

For now, Behal, Bricout and their team plan on repeating their study with a larger group of users to see if they can replicate their results. They’re also going back to users with MS, and perhaps traumatic brain injuries, early next year. Colleagues at other institutions are experimenting with the arms with even more diverse disabled populations.

The engineering team has already given the robotic arm a “voice” that announces its actions and makes it feel more responsive and less alien, even on automatic mode. They’re revamping the software interface again, including exploring the possibility of adding haptic feedback, so users can feel when the robotic arm can grasp an object — or the user’s body itself. If you’re going to scratch your head, the fingertips benefit from touch almost as much.

“You have to listen to users,” Behal said. “If they don’t like using the technology, they won’t. Then it doesn’t matter how well it does its job.”

Robotic arm’s big flaw: Patients say it’s ‘too easy’ [UCF Press Release]

See Also:

• Paging Dr. Robot
• Toyota Sees Robotic Nurses in Your Lonely Final Years
• Kid, Meet Your Old Man (and his Robot Body)
• Robo-Therapist Helps Ailing Limbs
• A Wheelchair That Reads Your Mind
• A Question of Mind Over Matter

How do you make a creepy baby robot downright cute? Give it an Indian headdress and teach it the bow-and-arrow, of course. The same team of researchers who brought us the pancake-flipping robot arm have imbued this iCub with a learning algorithm that lets it teach itself archery much the same as a human might do, by watching where the suction-tipped arrow lands and adjusting its aim for each subsequent shot. In this case, it obtained a perfect bullseye after just eight attempts. Watch it for yourself after the break, and ponder the fate of man — how can we possibly stop an uprising of adorable robots that never miss?

Continue reading Robot Archer iCub learns to shoot arrows, pierces our mortal heart (video)

Robot Archer iCub learns to shoot arrows, pierces our mortal heart (video) originally appeared on Engadget on Sat, 25 Sep 2010 18:28:00 EDT. Please see our terms for use of feeds.

Permalink   |  Petar Kormushev  | Email this | Comments

Marathons are not just for humans. A robot has decided to take on the challenge of walking 300 miles from Tokyo to Kyoto. The tour is part of a publicity tour organized for Panasonic’s Evolta batteries.

The 7-inch tall very cute humanoid robot that take on this project will be pulling a two-wheeled cart behind it. It’s a tiny machine but hopes to get to the finish line successfully.

The robot originally comes from the stable of Japanese company Robo Garage. It has been constructed using lightweight plastic, carbon fiber and titanium and weighs about 2.2 lbs. The entire machine will be powered using 12 AA batteries and operated using remote control, according to the Pink Tentacle site.

The robot will travel from sunrise to sunset, say the organizers, who will be tweeting its progress (@evoltatoukaidou) and livestreaming the event.

The robot can travel at a rate of about two to three miles an hour. So without any breakdowns or problems, the robot is expected to complete the journey in about 49 days.

It’s not the first time that this robot has undertaken adventure sports. In May 2008, it climed a 1740-ft rope suspended from a cliff at the Grand Canyon and a year later drove for a day around the Le Mans race circuit. All of this has already earned the robot a place in the Guinness World Records book.

For its current adventure, the robot has a wheel circling it so it can move over uneven surfaces. The handcart behind the robot is expected to hold extra batteries. The batteries will have to be recharged at least once every day.

Head over to the Pink Tentacle site to see photos of the very cute Evolta Panasonic robot as it gets ready to head out on its latest project.

Check out some photos of the Evolta from its earlier adventures:

The Evolta covered 14.8 miles at the Le Mans race circuit. Photo Courtesy Panasonic

Evolta robot's creator looks on proudly. Photo courtesy Panasonic

The Evolta robot has already set two records. Photo courtesy Panasonic

See Also:

• Adorable Walking Robot Sets Distance Record
• Robot Learns to Flip Pancakes
• Butler Robot Can Fetch Drinks, Snacks
• Anybots Robot Will Go to the Office for You
• Gallery: Robot Bartenders Sling Cocktails for Carbon-Based Forms
• Holographic Displays, Robot Eyes Hint at Your Interactive Future

Graphic by Christine Daniloff, via MIT News Office

Sometimes when one of my remotes is missing, I interrogate the others: “Where’s your friend? I know you know something!” In the future, with wireless positioning systems, a version of that method might actually almost work.

Researchers at MIT’s Wireless Communications and Network Sciences Group think networks of devices that communicate their positions to each other will work better than all of the devices transmitting to a single receiver. The latter is how GPS works, and if you’ve used it, you know it isn’t always very precise. In the lab, MIT’s robots can spot a wireless transmitter within a millimeter.

This seems almost intuitive: the more “eyes” you have on an object, the easier it is to triangulate — the robot version of “the wisdom of crowds.” But the key conceptual breakthrough here isn’t actually the number of transmitters or their network arrangement, but what they’re transmitting. MIT News’s Larry Hardesty writes:

Among [the research group’s] insights is that networks of wireless devices can improve the precision of their location estimates if they share information about their imprecision. Traditionally, a device broadcasting information about its location would simply offer up its best guess. But if, instead, it sent a probability distribution — a range of possible positions and their likelihood — the entire network would perform better as a whole. The problem is that sending the probability distribution requires more power and causes more interference than simply sending a guess, so it degrades the network’s performance. [The] group is currently working to understand the trade-off between broadcasting full-blown distributions and broadcasting sparser information about distributions.

Much of this research is still theoretical, or has only been deployed in lab settings. But Princeton’s H. Vincent Poor is optimistic about the MIT group’s approach: “I don’t see any major obstacles for transferring their basic research to practical applications. In fact, their research was motivated by the real-world need for high-accuracy location-awareness.” Like precisely which cushion my remote control is underneath.

Warning: Very Dry Flash Video Of Robots Finding Things Follows

See Also:

• Viruses Might Help Make Better Batteries
• Nike+ App Ditches Dongle, Gains GPS
• Man Scrawls World's Biggest Message With GPS 'Pen'
• Spime Watch: Shipping Containers and the Future Internet of Things …
• Spime Watch: why the Internet of Things is a fiasco
• New RFID Tag Could Mean the End of Bar Codes

Continue reading The state of household robots in Japan: looking pretty great

The state of household robots in Japan: looking pretty great originally appeared on Engadget on Sat, 04 Sep 2010 04:01:00 EDT. Please see our terms for use of feeds.

Permalink Technabob  |  Plastic Pals  | Email this | Comments