Forty-seven years ago this month, The Jetsons splashed onto our TV screens, promising us a world in which middle-class families would have Rosie The Robot - a lively, caring machine that could keep an eye on the kids, clean the house and scold the occasional rude guest.
We're not there yet but robot developers say we're getting closer.
"They'll drive cars, do your laundry, carry weapons and do bad things as well," said Jacky Baltes, a computer science professor at the University of Manitoba who has been developing a robot called Archie in conjunction with the Vienna University of Technology in Austria.
"This revolution is going to be much bigger than the computer revolution."
Prof. Baltes and his colleagues are aiming to eventually develop Archie into something very Jetsons-like - a robot that could assist the disabled, the elderly and everyday families in tasks around the home. Archie took part this summer in Robocup 2009 - a type of world championship for autonomous (non-remote controlled) robots.
For now, Archie can walk forward and backward. It's early days yet, but Prof. Baltes feels that Archie or some robot just like him will be fully functioning and affordable within a few decades.
The year 2050 is a target for many in the robotics world. It's when they hope to have an android capable of beating the world's best soccer player, much like the way a computer named Deep Blue beat chess champion Gary Kasparov in 1997.
Getting there will require huge leaps on two fronts. One, scientists will need to develop a robot that can move quickly and with agility on two feet. Secondly, robots will have to possess the type of artificial intelligence that will allow them to see their opponent, the ball and the field and make split-second decisions about which way to run or kick.
Toyota is already making strides on the first front. The Japanese company has made a robot that can run seven kilometres an hour on two legs that move very much like a human's. It can even stay balanced if given a push or shove. But, to date, it needs to operate in a controlled environment, on a flat surface, away from real-world obstacles.
A Rosie The Robot would need to get around furniture and would have to have supple hands capable of everything from grabbing a door handle to pouring a drink into a paper cup without any spills.
To do that, she would have to duplicate the complex actions of the human body. A simple act such as opening a door requires you to judge the distance to the doorknob, wrap your hand around it, and apply just the right amount of force to open the door without ripping it from its hinges. That same hand needs to be able to catch a ball, chop vegetables and negotiate the finer points of changing a diaper.
"Our muscles are really very complicated subgroups, so that individual components of one muscle can join with other components of another muscle to give a very subtle movement," said Alexander Ball, a professor of anatomy at McMaster University in Hamilton.
"So it's not like you've got two or three motors on one side of a limb that are able to ... give you a range of motion, but rather that there are hundreds and hundreds of these motors."
Prof. Baltes feels robots will become a lot more limber as new, smaller technologies come along to replace the various electric motors currently used to move robot limbs and joints.
"The material sciences are improving. People are working on artificial muscles," he said.
Progress is also being made on the second front. At Robocup and other competitions, teams of smaller robots play soccer. They can detect the ball, move toward it and kick or shove it toward the net, all without remote controls.
Robots also are getting better at wall-climbing competitions. At one event Prof. Baltes plans to attend, competitors this year are allowed to input the wall's map into the robot. But next year, the aim is to have robots capable of finding the route to the top of the wall, averting dead ends, all on their own.
In real-world applications, however, robots are still very limited in examining a situation and determining a response.
"A three-year-old can go into a room and say 'there's a chair and a door and there's mama and papa and all of that, but the best labs in the world can't write a computer program to come anywhere near that kind of environment recognition capability," Prof. Baltes said.
The huge gap between a soccer-playing robot and a multifunction, interactive Rosie does not deter Prof. Baltes, however. He chalks it up to a technology that is still in its infancy, much like airplanes a century ago.
"It's like looking at the Wright brothers in 1903 or '04 and they're flying 300 yards at Kitty Hawk, and then in 1947, Chuck Yeager breaks the sound barrier in a jet plane."
"These robots are getting better and better every year."Report Typo/Error