Spreading the Word on Robotics

It's one thing for mechatronics experts to cluster amongthemselves, sharing insights on their various disciplines. It's quiteanother for someone like Eric Gradman, a roboticist at a company calledApplied Minds in Glendale, CA to spread the word about his passion forrobotics. He spoke at an event called Mindshare in mid-October in LosAngeles, which brings together on a monthly basis an audience asdiverse as engineers, technologists, artists and dancers.

Of course, Gradman is not your ordinary roboticist. Sure, he has amaster's degree in robotics from the University of Southern California.He believes robotics encompasses mechatronics. "When I talk about why Ienjoy robotics so much, I say, when I'm confronted with a problem, onlyin robotics do I get the choice of solving it in hardware, software orelectronics."

But he's also an interactive artist and a circus performer.(Admittedly, roboticists and circus performers both employ a basicunderstanding of physics in their work).

Nor is Mindshare your ordinary event. According to one of thefounders, "We're intrigued by the cross pollination that arises fromgathering individuals who might not otherwise interact. It is anenergetic mix of open-minded people from architects to programmers towriters to painters to engineers and all others in between."

Clearly, Gradman, Mindshare and mechatronics all share a multidisciplinary view of the world.

Still, Gradman chose a unique perspective with which to discussrobotics with the Mindshare audience. His presentation was titled "ThePerils of Robotics: When Good Robots Do Bad Things." When he came upwith the title, he wanted to avoid giving a dry presentation aboutrobotics. "I love robotics," he says, "but it may have been that I washaving a particularly frustrating week at work, because I got tothinking about the many ways that robots can fail."

So why do good robots do bad things? "Robots can misbehave," saysGradman, who started out his presentation with the example of KenjiUrada, one of the first workers ever to be killed by a robot, "but theycan't be evil or stupid. It comes down to a human failure to designaround a problem. Robots fail because we don't know how to solve allthe problems (in robotics). We don't understand how to make themsucceed yet."

Gradman says there are multiple ways robots foul up basic activities they're supposed to understand, including:

Sensing their environment. When robots use lasersas sensors to understand their environment, they can be fooled byreflective materials. The laser can be diffused to a point where itcan't recognize an object. This is fine, Gradman says, except veryexpensive sports cars tend to have exactly the kind of paint job thatfool the lasers. "In a parking lot full of cars, the robot will runinto the most expensive one," says Gradman. "I think that in thefuture, to avoid hostile robots, people will wear clothing made ofsilver lamé and mylar."

Modeling their environment. Gradman worksfrequently with computer vision, so he's come to understand robots'limitations in this area. In his presentation, he showed examples of arobot accurately tracking the face of a woman who's moving around aroom. But he also shows an example of a robot using standard trackingalgorithms that is fooled into thinking a smiley face painted on ahuman hand is another human face.

Planning a course of action. Sometimes engineersask robots to take an action that requires too much flexibility in toosmall a space or too short a time, Gradman says. "All of a sudden therobot is asked to do something that's both fast and impossible, andwithout safeguards you could put people in danger." And sometimesrobots just don't do what they're supposed to.

Acting to achieve the goal. And sometimes there's just hardware failure. Gradman included in his presentation videoof Honda's Asimo robot ascending a set of stairs, turning to descend asecond set of stairs, and toppling forward catastrophically.

Why are robots prone to failure? The answer is simple, in Gradman'seyes. "Robotics is still about human beings putting together robotsystems. We become part of the robot for a brief period of time, andidiots can cause more damage through robots than any robot can byitself."

For all his humorous examples of robotics failure, Gradman is highlyoptimistic about the future of robotics. In September, he attended IROS2008 (Intelligent Robots and Systems) in Nice, France. "It's very clearfrom listening to all the talks there that we are rocketing ahead instate-of-the-art sensing, planning and modeling," Gradman says."Processing power is increasing, so we can build more nuancedalgorithms and more sophisticated mathematical techniques to deal withinteresting exceptions in data. Sensors are getting better, and ourmethods of dealing with the sensor data are getting better. Brilliantresearchers are making great inroads in all these things, so we're onthe path to solving a lot of the problems I talked about."

Spreading the Word on Robotics_ImageA