Perfect Helper: Is a Robot in Your Future?

© 2007 iRobot Corporation. All Rights Reserved.

Like many active 68-year-old retirees, Lavonne Knapp sometimes overdoes it in the garden. But when her back is killing her after a good bout of weeding—and, heck, even when it isn’t—she simply has her robots clean her floors.

Knapp owns both a Roomba and a Scooba, the squat, disc-shaped vacuum and scrubber that automatically navigate her home with no more effort on her part than pushing the “on” button. “How could you not love that?” asks the Roseburg, Ore., native. “They’re marvelous, and there’s nothing to it.”

If only it was so easy to deliver on the larger promise of the all-purpose robotic aide that scientists, futurists and the media keep trumpeting. At least once a year, it seems, a new experimental robot for the home is heralded as the imminent savior of older Americans and those with disabilities. Indeed, scientists at the University of Massachusetts Amherst recently showed off the latest example: the uBot-5, which can dial 9-1-1, provide reminders (to take medication or turn off the stove), lift packages and more.

But how soon will it and similar robots become marketable—when, really, might they start lending a hand at home?

The need is certainly real: The aging population is exploding worldwide, resulting in more people who are living on their own but are unable to perform daily activities unassisted. Meanwhile, health care costs, including nursing homes and home-care nurses, continue to skyrocket.

These trends merge to pose a challenge: How can society affordably increase the comfort of more and more older Americans who want to remain independent with the requisite level of dignity?

Some people clearly see robots as the answer, or part of it. Honda and Toyota have both been racing to produce “partner robots” that are roughly shaped like humans and can help people perform tasks, such as walking, climbing stairs and carrying stuff. Carnegie Mellon University in Pittsburgh is developing a “nursebot” called Flo (for Florence Nightingale), and a postdoctoral student at M.I.T.’s computer science and artificial intelligence lab is developing a robot called Domo that can identify groceries by shaking them in its hand—before putting them away in a cupboard.

Still Working Out The Bugs
Though certainly impressive, these robots won’t be leaving research and development labs anytime soon. And, some say, they may never graduate.

“Honestly speaking, I don’t think robotics is ready for the robot-as-human-helper,” says Takeo Kanade, a professor of robotics at Carnegie Mellon and director of the school’s quality of life technology center.

Kanade recognizes how useful it would be to have a droid that can perform specific tasks, even something as simple as reminding you to turn off the stove. But he suspects that all-purpose anthropomorphic robots are still the stuff of science fiction.

Honda’s ASIMO, for instance, is an astoundingly realistic walker and can be programmed, among other things, to conduct the Detroit Symphony Orchestra through one song, as it did last month. Achieving the simple manipulation of objects is relatively easy, Kanade points out. But getting, say, a kitchen-based robot to differentiate between the layout of your kitchen and that of your neighbor’s—to recognize which of your dishes are dirty, clear your table, load your dishwasher and not break anything—is still an unrealized dream.

Machines For Living In
Instead, Kanade is more inclined to envision a world where people live inside robots, rather than with them: “I would summarize [the future] as house-as-robot.” Our cars are already getting smarter, he points out: They’re equipped with GPS (Global Positioning Systems) and can help us parallel park. For your home, you can get surveillance systems or floors that can tell if a person has fallen. He foresees buildings that can detect if water has been left running or a stove has been left on—and then address the problem.

Young or old, who wouldn’t want to live with these features?

“I like to call it invisible robotics,” he says. In Kanade’s world, robots might someday lift the odd package or remind you to take your heart pills. But the main purpose of robotics is to improve your overall environment, floor to ceiling. A dawn-to-dusk buddy that looks like R2D2? Forget about it.

Kanade’s colleague at Carnegie Mellon, Jim Osborn, is a big advocate of thinking small. The goal of quality-of-life robotics, he says, should be to understand when somebody needs a reminder, or help through a sequence of steps in performing tasks. His team is working on technology that would comprise a camera and an earpiece, and not be any bigger than a lapel pin.

“It would know all the environments that you’re in and remember all of the scenery in your daily life,” he says. “It could have facial recognition to help remind you of people’s names. And based not just on the time of day but based on what you’ve actually done, remind you to, say, take your medicine—and more importantly, not remind you to take your medicine if it knows you already have.”

His point, he says, is that more important than a pie-in-the-sky humanlike android is the development of robotics that help older Americans to live independently. He calls it “robot-person symbiosis.”

© 2007 iRobot Corporation. All Rights Reserved.

As Lavonne Knapp and her Roomba attest, symbiosis isn’t that hard to achieve, even today, if the expectations for what robots can provide are scaled down—way down—from their C3PO heights. Indeed, over the course of about six years, the number of helper robots in the world has gone from essentially zero to 6 million. And the majority of them are made by iRobot, the Boston-based company that manufactures the Roomba and Scooba, ranging in price from $200-$350. (iRobot has also shipped more than a thousand bomb-retrieving robots to Iraq and Afghanistan).

“Products like the Roomba have become mainstream,” says Philip Solis of ABI Research, a technology market analysis firm based in Oyster Bay, N.Y. “What we see today are single-task robots that are relatively low-cost. It’s really mostly vaccuums, floor washers, a pool cleaner and lawn mowers.” Solis estimates the single-task robot marketplace today to be worth $154.7 million but ballooning to $8.7 billion by 2015.

Starter Robots
This projected growth has not escaped iRobot CEO Colin Angle, who says that the Roomba and Scooba are getting boomers used to the idea of living with robots. “As the boomers start to require more care, we have a real challenge on our hands,” he says. “The solution is to let people live independently where they currently live, for longer periods of time. The robots that we’re developing are aimed at trying to make that possible. It’s going to be an incremental thing, and we’ve started by helping maintain the home.”

To that end, the company has a new helper robot entering the beta phase of its development. The ConnectR Virtual Visiting Robot, which will cost between $500 and $600, is armed with a Web cam: It would live in your house, and anybody with the correct security code could log into the robot to visit you virtually, through an LCD monitor. It could also be used to routinely stay in touch and be sure a parent is staying on top of medication. Or a nurse or doctor could make a virtual house call by dialing in to the robot.

The medical field is particularly fertile territory for exploring the potential of robotics. A company called InTouch Health in Santa Barbara, Calif., has deployed primitive-looking robots to several hospital emergency rooms and nursing homes around the country for a couple of years. Piloted remotely by doctors and residents who help diagnose patients through video camera interviews, these robots enhance convenience for physicians and give patients quicker access to doctors who aren’t in the hospital.

Now a Boston start-up robotics company called Myomo has been granted approval by the U.S. Food and Drug Administration for a robotic elbow for the rehabilitation of stroke patients. The small device is strapped to a patient’s bicep where it picks up weak flickers of movement. It sends that signal to a one-pound box worn on the person’s body. From the box, the signal is magnified, allowing the person to move her arm.

“Now the person can carry groceries, the laundry basket or can even walk because she can use a cane again,” says Ela Lewis, director of product management at Myomo. “Our goal is to empower the person at home to be more independent with robotics.”

Looking into the future, it’s safe to expect small, single-function robots—more like electronic gadgets or even new features of existing devices, such as the parallel parking aids in some cars today—to continue to improve our lives, says Kanade. Further down the road, you can look for some “partner robots” to arrive.

Japan is now racing far ahead of other countries in developing robots with more human features—or robots that can interact more easily with people. The best of these will be able to perform multiple simple tasks in caring for Japan’s sick—or collecting the trash, guarding homes and offices, and giving directions on the street—by 2015, Kanade says. Still, the likely cost (tens of thousands of dollars) may prove prohibitive for many would-be early adopters.

Whatever shape they end up taking, when those first intelligent partner robots of the future do arrive, one of their programmed tasks ought to be thanking their older Roomba cousin for keeping the floors clean.

Brian Braiker is a general editor at Newsweek magazine, where he writes frequently about technology and popular culture.