Australians and New Zealanders have landed in Korea to contest the 2002 FIRA (Federation of International Robot-soccer Association) Robot Soccer World Championship, which began in Seoul today.

-In the long term, the goal of RoboCup and FIRA is to develop a team of robots that can beat the human world champion in a game of soccer. The deadline for this goal is 2050," Hansjoerg (Jacky) Baltes, senior lecturer, Department of Computer Science, University of Auckland, told ZDNet Australia via e-mail. His students are competing with two teams in the competition.

According to Dr Joaquin Sitte, head of the Smart Devices Lab at Queensland University of Technology (QUT), the idea behind robotic soccer is to test robotics, artificial intelligence and visual systems in a challenging real-time environment. The FIRA competition distinguishes itself from other competitions, such as RoboCup, by having a restrictive set of rules, forcing players to differentiate themselves via the strategy they use rather than losing to the team with the best hardware.

-It's an environment which is manageable, but is still very rich in its problems. It's good for teaching, but also stretches to research," Sitte told ZDNet Australia. Sitte heads up the QUT team, which will compete in two categories.

The competition has 10 different games, each with different specifications for the robots. The robot players are radio controlled from a central computer, which obtains visual data of the playing field via a camera in the ceiling.

-This allows the processing capacity to be bigger than what you could put on the robots," said Sitte, adding that eventually the robots will be completely autonomous. Once the game is underway no humans are allowed to do anything.

Australia and New Zealand have several teams entering the competition, which sees competitors come from all over the world to pit themselves against each other. Most teams are from universities because of the huge investment in time and money required.

-It's good for students to learn about real-time computing systems," according to Sitte. -Everything has to happen at the speed of the game. They have to come up with things that work in reality, not just computer simulations, and compete against other teams, who are smart too."

Baltes agreed and also emphasised the unique environment. -You have 300 people working basically 24 hours before the competition. This included new students as well as some of the most important names in Artificial Intelligence research. For example, Drew McDermott, who is a really well known professor, was working pretty much non-stop. It is an electric atmosphere. There is intense passion by all of those involved."

The competition often pushes the boundaries; with teams sometimes getting feedback that what they've done is impossible, according to Baltes. -It's nice to be able to show an existence proof." This feeds naturally into many real-life research projects. One such application for the technology is Urban Search and Rescue, which Baltes is researching.

Sitte also sees the scientific value of the competition. -There aren't any great leaps, but there are some incremental results in many areas, particularly the field of vision," he said. -It's quite a challenging task which stretches our current computational algorithms." He points to participation in earlier competitions of companies who specialised in frame grabbing, a type of vision hardware.

So how long, realistically, before these advances are available for the man on the street? -We are at the stage where the personal computer was 20 years ago, where it really started with the hobbyist," Sitte said. -That mirrors these developments. We are also seeing some commercial products, some companies are selling small robots, as toys."

Copyright © 2009 CBS Interactive, a CBS Company. All Rights Reserved.
ZDNET is a registered service mark of CBS Interactive. ZDNET Logo is a service mark of CBS Interactive.