Reading Robot Learns Like A Human

Meet “Marge”, a young robot that knows that Barclays is a UK-based bank and is smart enough to read the New York Times. Thanks to artificial intuition, teaching a robot how to read and comprehend symbols has been achieved by robotic researchers. How is this possible?  In order to create a robot that could read almost as well as a human, Ingmar Posner and Paul Newman at the University of Oxford and collaborator Peter Corke at Queensland University of Technology had to address what allowed humans to read text better than robots. Edward Grant, Director of the Center for Robotics and Intelligent Machines at North Carolina State University explains:

To a literate human, reading is a simple matter. If a word changes size, or the lighting in the room changes, you don’t instantly become illiterate. That’s because the human brain can make intuitive leaps of logic. The brain says, ‘I’ve done something kinda like this before, so I can adapt to this new activity that has been presented to me.’  A robot, or more specifically, the mathematical algorithms installed on a robot that are its brain, can’t make that intuitive leap. For a robot, a change in the angle, lighting or size means they have to learn to read all over again.

Robots generally cannot make such leaps in intuition, but in Marge’s case, her creators installed text recognition software (Optical Character Recognition), a spell-checker, and a dictionary into her. By implementing some new software tricks to make sure she is reading text and not random cracks in the sidewalk, Marge is able to determine a word’s meaning, effectively bridging the gap in intuition. Since Marge’s intelligence comes from software, not hardware, the potential for cellphones and even eyewear can be explored.

The Oxford Mobile Robotics Group

Robotics, Vision and Sensor Networking at QUT

[via Discovery News]