Robot Kits
Stiquito Kit
BattleKits
Robot Toys
Solar Kits
Robot Arms
Robosapien
Basic Stamp Kits
Lego MindStorms
Beginners
Books
Hobby Robots
Robot Sports
Electronics
Mechanics
Robot Minds
Books for Kids
Robots at Work
Microcontrollers
Advanced Books
Used Books
Real Robots
Robot Motors
Remote Controls
Robot Parts
Robot Tools
Magazines
Robot Videos
Robot News
RoboLinks
Contact
|
Courtesy of New Scientist Magazine
By Duncan Graham-Rowe
A DISEMBODIED HEAD with gremlin-like features is about to be given life in the artificial
intelligence department of the Massachusetts Institute of Technology. Named Kismet, the
head is a robot that learns about its environment like a baby, depending almost entirely
upon benevolent caregivers to help it find out about the world.
Researchers working on Kismet are part of a larger group in the AI department that is
building an artificial humanoid called Cog. Kismet is Cog's baby brother, and what the
researchers learn while putting the robots together will be shared to speed up
development.
Once finished Cog will have everything except legs, whereas Kismet has only a 3.6-kilogram
head that can display a wide variety of emotions. To do this Kismet has been given movable
facial features that can express basic emotional states that resemble those of a human
infant. Kismet can thus let its "parents" know whether it needs more or less
stimulation-an interactive process that the researchers hope will produce an intelligent
robot that has some basic "understanding" of the world.
This approach of creating AI by building on basic behaviors through interactive learning
contrasts with older methods, in which a computer is loaded with lots of facts about the
world in the hope that intelligence will eventually emerge.
Cynthia Breazeal, who created Kismet, is giving the robot the same senses as a baby. So
far it has only a basic vision system, but Breazeal is developing ways to make Kismet hear
and babble like a baby. Its eyes are made of charge-coupled device color cameras that
focus best at a distance of 0.6 meters-a visual acuity similar to that of a baby.
Kismet has a repertoire of responses driven by emotive and behavioral systems. The hope is
that Kismet will be able to build upon these basic responses after it is switched on or
"born", learn all about the world and become intelligent.
Crucial to its drives are the behaviors that Kismet uses to keep its emotional balance.
For example, when there are no visual cues to stimulate it, such as a face or toy, it will
become increasingly sad and lonely and look for people to play with.
Responding to Kismet restores its equilibrium, making it happy again. Similarly, if its
caregiver endlessly repeats the same cue, such as shaking a doll in front of it, it will
get bored and agitated. And if Kismet becomes overwhelmed with information, it is likely
to tire and fall asleep.
These responses are crucial to the project. Social interaction is always a two-way
process: Kismet must be able to show its parents how it feels, so they can respond
correctly. The robotic face sports ears, eyebrows, eyelids and a mouth that can move to
form baby-like facial expressions such as surprise, fear, interest and so on. "The
argument is that many infants have this innate ability so that if you overstimulate or
understimulate it, the infant lets you know," says Breazeal. She suggests that when
parents respond consistently to an infant they are creating a rich environment for the
baby to learn how the social world works, as well as a foundation for its own personality
to develop.
Any advances made with Kismet will be passed on to its big brother Cog, the robot
brainchild of Rodney Brooks, head of MIT's AI department. Cog is 2 meters tall, complete
with arms, hands and all three senses-including touch-sensitive skin. Its makers will
eventually try to use the same sort of social interaction as Kismet to help Cog develop
intelligence equivalent to that of a two-year-old child.
Kismet is still very much in development, with Breazeal still tuning its emotive responses
and working on its sensory apparatus. But she expects learning to start within a year.
Will it work? "No one knows what the answer is because no one has done it yet,"
says Breazeal. "The only existing proof of intelligence is these walking systems in
our world called humans and animals. So the case for embodiment is that if we do it right
it will probably lead us where we want."
Meanwhile, Cog is getting close to completion. Brooks says all its components will be
integrated over the next few months and once they are, he speculates that it can be
considered to be an artificial life form.
Subscribe to New Scientist
|
|