A Brief History of RoboSoccer or A Standard Problem of Artificial Intelligence

Right from the start, work in the fields of robotics and artificial intelligence (AI) has revolved around the emulation of human capabilities. This is made particularly clear by the robot’s linguistic origins: derived from the Czech word “robot,” the concept of mechanical “corvée or forced labor” (link to http://en.wikipedia.org/wiki/Robot) first appeared in the early 20th century in the science fiction works of the Capek Brothers. Josef Capek, in his 1921 play “R.U.R.,” described anthropoid laborers bred in tanks (whereby the author had recourse to the Golem motif). Then, in 1942, Isaac Asimov (link to http://en.wikipedia.org/wiki/Isaac_Asimov) formulated his famous Three Laws of Robotics in his short story “Runaround”:

1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. A robot must obey orders given to it by human beings, except where such orders would conflict with the First Law.
3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Law.

An extremely prolific author, Asimov also wrote the short story that was the basis of “I, Robot,” the 2004 film starring Will Smith.

It was George C. Devol’s 1954 patent application for a programmable manipulator that announced the birth of the world’s first actual robot. Devol was a co-founder of the Unimation Company that introduced the first hydraulic industrial robot in 1960. Eight years later, the Massachusetts Institute of Technology (MIT) (link to http://en.wikipedia.org/wiki/MIT) in Cambridge USA developed the first mobile robot.

Ever since the mid-20th century, scientists (equipped with ever more powerful computers) have been striving to surpass the capabilities of human intelligence.

[NOTE: In an address on the subject of the future of mechanical intelligence delivered at a meeting of the American Association for the Advancement of Science (AAAS), computer expert Ray Kurzweil expressed the opinion that AI will equal and even surpass human intelligence within the next 30 years.]

But, let’s return to the 20th century for now. By far the most famous domain of competition between man and machine is (computer) chess (link to http://en.wikipedia.org/wiki/Computer_chess). Indeed, progress in this field has been much slower than originally projected, but already in 1996, the chess program DEEP BLUE succeeded in winning two games against then-world champion Garri Kasparov. Nevertheless, even before this initial triumph, informed observers were coming to the conclusion that machines, in order to achieve parity with human beings, had to be able to do more than just think fast. True AI means being able to take action as well!

The state of development of the various manipulative devices that were then in use in many different areas of industrial production was already quite amazing. Thus, it’s no wonder that the demands for more intensive exploitation of this potential was becoming louder and louder. What these advocates were calling for was to get these “robots” out of their prescribed operational radii, fixed tracks and limited ranges of motion and to outfit them with AI to enable them to find their own way in unfamiliar territory.

A process of rethinking commenced in the early ‘90s. For the robots of the future, being provided (in advance) with precise information about their environment would not be as important as their ability to perceive it themselves and react to it. Rodney A. Brooks’ (link to http://en.wikipedia.org/wiki/Rodney_A._Brooks) considerations on the subject of behavior-based robotics in his famous 1990 article “Elephant’s Don’t Play Chess” highlighted the moment of the paradigm shift to robots that can perceive their environment instead of already having to know it.

Then in 1993, Canadian Alan Mackworth (link to http://en.wikipedia.org/wiki/Alan_Mackworth) suggested making a football match the standard problem in the field of AI. After all, this one task encompasses the entire spectrum of relevant R&D efforts: sensorics, robotics, artificial perception, drawing logical conclusions, image processing, multi-agent systems, etc. etc. And the icing on the cake is that everything takes place under real-time conditions. Plus, the sporting rivalry that’s inherent in football brings an additional decisive advantage: football tournaments are a great way to foster friendly competition that enables the most successful solutions to achieve a breakthrough. This, in turn, advances the development of the whole field of robotics.

In 1995, Korea’s Jong-Hwan Kim launched the international organizing committee of the Micro-Robot World Cup Soccer Tournament. The first international MiroSot championship took place in 1996 at the Korea Advanced Institute of Science and Technology (KAIST) in Daejeon, South Korea. One year later, FIRA–Federation of International Robot-Soccer Association (link to www.fira.net) was organized. Fans of robot football were getting geared up in Japan at this time as well. In 1997, the first RoboCup Competitions (link to www.robocup.org) took place. Since the first world championship in 1998 in France (FIRA Cup and RoboCup), world and European championships in robot football have been held annually.