Good lord Chris, those are very impressive. My first thought wasn't how cool they would look skinned in a game, but how powerful this is to the field of robotics. Build your real robots as models in this gravity/friction environment and use GAs to converge on behaviour (that you could never have time to train IRL) then pop the results back into your real bot.
I love it when the narrator says:
"Perhaps it is mean to move the goal away just as it arrives"
You bet matey, that things gonna bite your ass when it evolves teeth.
On Mon, 12 Nov 2007 01:03:58 -0500 Chris McCormick chris@mccormick.cx wrote:
On Sat, Nov 10, 2007 at 02:45:52PM +0000, Andy Farnell wrote:
Or when you say "graphics" do you mean something else. I'm having a hard time imagining any kind of use for procedural geometry other than for liquids and gasses (fire and water again) because non-living solids tend not to change their topology. The exception would be trees and plants built using the L-system type bifurcation I guess.
There's a talk I've given at a couple of conferences on using procedural algorithms in games, and when I get to the genetic algorithms bit I always show this video, which pretty much single-handedly convinces the entire class of the validity of what I'm saying:
http://www.genarts.com/karl/evolved-virtual-creatures.html
1994! It still gives me goose bumps.
Best,
Chris.