The Do-Anything Robot
Your PC can accomplish any computing task you ask of it. Why isn’t the same true for robots
By Gregory Mone
Robots have proved to be valuable tools for soldiers, surgeons and homeowners hoping to keep the carpet clean. But in each case, they are designed and built specifically for the job. Now there is a movement under way to build multipurpose machines—robots that can navigate changing environments such as offices or living rooms and work with their hands.
All-purpose robots are not, of course, a new vision. “It’s been five or 10 years from happening for about 50 years,” says Eric Berger, co-director of the Personal Robotics Program at Willow Garage, a Silicon Valley start-up. The delay is in part because even simple tasks require a huge set of capabilities. For a robot to fetch a mug, for example, it needs to make sense of data gathered by a variety of sensors—laser scanners identifying potential obstacles, cameras searching for the target, force feedback in the fingers that grasp the mug, and more. Yet Berger and other experts are confident that real progress could be made in the next decade.
The problem, according to Willow Garage, is the lack of a common platform for all that computational effort. Instead of building on the capabilities of a single machine, everyone is designing robots, and the software to control them, from the ground up. To help change this, Willow Garage is currently producing 25 copies of its model PR2 (for “Personal Robot 2”), a two-armed, wheeled machine that can unplug an appliance, open doors and move through a room. Ten of the robots will stay in-house, but 10 more will go to outside research groups, and everyone will pool their advances. This way, Berger says, if you want to build the robotic equivalent of a Twitter, you won’t start by constructing a computer: “you build the thing that’s new.”
The military, short on linguists, is building smart phone–based devices to do the job
By Gregory Mone
Sakhr Software, a company that builds automatic language translators, recently unveiled a prototype smart phone application that transforms spoken English phrases into spoken Arabic, and vice versa, in near real time. The technology isn’t quite ready for your next trip to Cairo, but thanks to recent advances in machine-translation techniques, plus the advent of higher-fidelity microphones and increasing processing power in smart phones, this mobile technology could soon allow two people speaking different languages to have basic conversations.
Before the 1990s automatic translation meant programming in an endless list of linguistic rules, a technique that proved too labor-intensive and insufficiently accurate. Today’s leading programs—developed by BBN Technologies, IBM, Sakhr and others as part of a Defense Advanced Research Projects Agency effort to eliminate the military’s need for human translators—rely on machine-learning techniques instead. The software works from a database of parallel texts—for example, War and Peace in two different languages, translated United Nations speeches, and documents pulled off the Web. Algorithms identify short matching phrases across sources, and the software uses them to build statistical models that link English phrases to Arabic ones.
John Makhoul, BBN’s chief scientist, says the current technology is at its best when confined to subject areas with specific phrases and terminology—translating a weather report from English into French, for example, or helping soldiers gather basic biographical information from people in the field. Makhoul envisions the first consumer applications, five years from now, being similarly constrained. A tourism-related translation app on a smart phone could help an American in Florence get directions from a non-English-speaking local, but they won’t chat about Renaissance art. “It is not going to work perfectly,” he says, “but it will do a pretty good job.”