When Robots Play Games

Roland Piquepaille writes "If the theory of evolution has worked well for us -- even if this is arguable these days -- why not apply it to mobile robots?, asks Technology Research News. Several U.S. researchers just did that and trained neural networks to play the Capture the flag game. Once the neural networks were good enough at the game, they transferred them to the robots' onboard computers. These teams of mobile robots, named EvBots (for Evolution Robots), were then also able to play the game successfully. This method could be used to build environment-aware autonomous robots able to clear a minefield or find heat sources in a collapsed building within 3 to 6 years. But the researchers want to build controllers for robots that adapt to completely unknown environments. And this will not happen before 10 or maybe 50 years. You'll find more details and references in this overview, including a picture of EvBots trying to find their way during a game." Read on for a similar robot competition held this weekend in France.

saunabad writes "The annual Eurobot autonomous robot contest for amateurs is held this weekend on La Férte-Bernard, France. This year's theme is 'coconut rugby,' and the robots are collecting small stress balls from the field and carrying them to the opponent's end, or shooting them in the rugby goal, while avoiding the randomly placed obstacles at the same time. Each team has a one main robot and an optional small assisting robot."

Bzflag

[super-momo]

Wikipedia mentioned Quake and UT. Bzflag [bzflag.org] is also a great CTF game, and a classic.

Re:"even if this is arguable these days"?

[Pluvius]

It's obviously a reference to how stupid people seem to be these days. Of course, evolution only enhances intelligence when intelligence is required to better survive and procreate--but it's a joke, you're not supposed to read that much into it anyway.

Rob

neural nets != genetic algorithms

[Tablizer]

The introduction makes it sound like training neural networks is evolution. Neural networks and genetic algorithms are two very different technologies, although they can be combined.

Re:Arguable?

[tkittel]

> There are no privilaged inertial frames, but the
> Earth is not in an inertial frame. If it was, it
> would be shooting off in a straight line at a
> constant velocity.

Actually "straight line" is also undefined in this context. A frame of reference attached to the earth, is indeed not an intertial frame. But the way to see this is because the laws of Newton are invalid if expressed in the coordinates of such a frame (even in the unrelativistic limit).

Re:Hubris

[Tyreth]

Though I reject darwinism, natural selection and adaptaion still play a role. They could potentially be quite useful for the creation of robots as suggested.

Our God is a Creator God, and He made us in His image. We therefore are also creators - mimicking Him. If our creations are a rebellion against Him, then of course He will be displeased. But it is possible for us to do these things and still give Him honour. The tower of Babel was a direct attack on God:
Come, let us bild ourselves a city, and a tower whose top is in the heavens; let us make a name for ourselves, lest we be scattered abroad over the face of the whole earth.

It sounds to me like this was a plan directly aimed as a defense against God should He choose to intervene.

Re:neural nets != genetic algorithms

[DaJoky]

Absolutly right, according to the article on the Evolution Train Robot Teams [trnmag.com] it is the combination of both method : "After several hundred generations, the neural networks had evolved well enough to play the game competently and were transferred into real robots for testing in a real environment."