Robots invent spoken language, join Facebook

May 18, 2011 by Amara D. Angelica

OK, I just made up the Facebook part, but IEEE Spectrum reported Tuesday on two robots that communicate linguistically like humans and invent new words. Spooky.

They’re called “Lingodroids” (reminds me of Stephen King’s even spookier The Langoliers, which were robotic monsters dealing with a “time rip”).

(credit: Ruth Schulz)

Researchers at the University of Queensland and Queensland University of Technology in Australia have designed the Lingodroids as a mobile platform equipped with camera, laser range finder, and sonar for mapping and obstacle avoidance. The robots also carry a microphone and speakers to “talk” to each other.

If one of the robots finds itself in an unfamiliar area, it’ll make up a word to describe it, choosing a random combination from a set of syllables. It then communicates that word to the other robot. (Sounds like what children often do.)

Lingo Droid

A game allows the Lingodroid robots (A) to develop and test their lexicon by specifying a target location (B) to meet at (credit: Ruth Schulz)

The robots then play games with each other to reinforce the language. For example, one robot might say “kuzo,” and then, both robots will race to where they think “kuzo” is. When they meet at the same place, that reinforces the connection between a word and a location.

In the future, researchers hope to enable the Lingodroids to “talk” about even more elaborate concepts, like descriptions of how to get to a place or the accessibility of places on the map, Spectrum points out. Ultimately, techniques like this may help robots to communicate with each other more effectively, and may even enable novel ways for robots to talk to humans.

Social networking for robots?

Let’s take it a step further. How about turning these bots (actually, many bots) loose in some kind of rich environment, like a simulated mall? Then you could connect them up to Amazon’s Mechanical Turk service, where people could remotely see what the robots are doing and help them learn (or some will mess with their minds).

Once these bots got really smart (and Asimov’s three laws of robotics, plus other laws, were drilled into them), you might turn them loose in the real world. Imagine bots driving around in autonomous Google cars, delivering pizza.

OK, now I’m scaring myself.

Eventually, the bots could evolve their own social life, keeping each other informed and writing silly little notes to each other. (Would they then be facebots?) They could also add information from the real world (using “Internet of things” data) and have their own IP numbers and email addresses. (Hmm, probably a bad idea: think spambots.)

It’s a small step from there to our facebots replacing DMV clerks, consumer help operators, McDonald’s counter workers (“DO.YOU.WANT.CHIPS.WITH.THAT.BURGER?”) and website editors.

Sorry I brought it up.

Amara D. Angelica is Editor of KurzweilAI