/ 14 April 2008

Robots, our new friends electric?

Fictional robots always have a personality: Marvin was paranoid, C-3PO was fussy and HAL 9000 was murderous. But reality is disappointingly different. Sophisticated enough to assemble cars and assist during complex surgery, modern robots are dumb automatons, incapable of striking up relationships with their human operators.

But that could soon change. Engineers argue that, as robots begin to form a bigger part of society, the new machines will need a way to interact with humans. In short, they will need artificial personalities.

This week, engineers, psychologists and computer scientists from across Europe will begin a major project that aims to develop the first robot personalities.

”What we’re looking at here is long-term interactions between people and robots in real situations,” said Peter McOwan of Queen Mary, University of London, coordinator of the £6,6-million, European Union-funded Lirec project.

”The big question is: What sort of properties does a synthetic companion need to have so that you feel you want to engage in a relationship with it over an extended period of time?”

Lirec — Living with Robots and Interactive Companions — consists of 10 university partners from seven countries that will run for just more than four years.

Phones and computers have shown how people can develop relationships with inanimate electronic objects. The next generation of digital servants will deepen these relationships.

In future, McOwan imagines robots as helpers around the house, acting as companions or integrating with the web to order groceries online.

”We’re also looking at support for the elderly,” he said. ”One of the projects is the ‘spirit of the house’, where there’s an entity monitoring to make sure no one’s fallen over, that they’ve taken their pills.”

Giving the ”spirit of the house” a human-like persona, by using sophisticated programs that can learn their users’ preferences, would make it more trustworthy for people interacting with it. ”You are interacting with a digital entity that is far more naturalistic than sitting with a keyboard and a mouse,” he said.

Kerstin Dautenhahn, a professor of artificial intelligence at the University of Hertfordshire, has developed Kaspar, a robot in the shape of a two-year-old boy, which can make facial expressions and play games such as peek-a-boo.

Dautenhahn has also set up a flat in Hatfield, where a home-help robot interacts with volunteers, to study longer-term relationships between people and machines. Research has found that the look of a robot should depend on the person it is mostly interacting with.

”People who are more extroverted say they enjoy interacting with a humanoid robot,” said Dautenhahn. ”People who are more introverted are more comfortable with a more mechanical robot.”

McOwan is confident people will be willing to form relationships with the robots. ”We anthropomorphise all over the place when it comes to interacting with computers — they’re either spiteful or friendly or whatever.”

The key will be to replicate the verbal and non-verbal cues people use to communicate. ”What we hope to do is produce something within which you feel there is something you’re bonding with.”

Dream machines

1495 — Leonardo da Vinci has the idea of a ”mechanical knight”.

1738 — French inventor Jacques de Vaucanson builds a mechanical duck.

1921 — The word ”robot” coined by Czech artist Josef Capek, from the Czech word for ”compulsory labour”.

1950s — Unimation introduces first commercial robots (for car factories).

1969 — Victor Scheinman, a Stanford AI lab student, creates the Stanford Arm, a predecessor of all robot arms.

1998 — The Furby robot is a bestseller.

1999 — Sony releases its Aibo robot pet.

2000 — Honda introduces its humanoid Asimo robot.

2003 — Nasa launches robotic Mars explorers, Spirit and Opportunity.

— guardian.co.uk Â