Heck, worth a vist just to see my noble visage (not to mention Amit's)...
In case you're thinking this might be "just another GA toolkit", you should know that shat caught my interest was an article which talked about researchers discovering digital organisms which had evolved to be smarter, grow larger, and even play dead when the researchers "killed" them...fascinating stuff.
Very interesting stuff...a good developer might be able to really build something using this as a base...
AI aspects of the game, besides the monsters communicating on your location and learning it from one another, include an A-Life component in which the monsters can spawn baby monsters if sufficient numbers of them get together. Left to themselves the new monsters can quickly outmatch the player's ability to fight them, so the player is forced to move against them and try to break their spawning groups up.
The monsters themselves apparently use genetic algorithms between games to slowly evolve strategies of movement and search to engage the player. This makes the game somewhat like CDDNA which also uses GAs for similar purposes. The monsters will slowly adapt to the player's style of play over time, which can make for some formidable opponents.
As mentioned earlier the whole game is available under the GNU license, so you can download and play with it yourself if you want. The code is of course freely available as well for tinkerers (which is why I've put this into the categoy of ExAI games as well).
If you want to give this puppy a try, hop over to the site and give it a shot....
Recommended.
In an interview in EE Times, Ariel described the toolkit as "...a general-purpose toolkit implemented in Java that I designed for experimenting with genetic algorithms as solutions to optimization problems." The kit allow users to define fitness functions and parameters in plain text, then kick off a run to find a solution for the problem at hand. Tools are provided to monitor successive generations of solutions, and the user can interrupt the process to make changes directly at any time.
I only fiddled with this toolkit for a few hours and I barely scratched the surface. If genetic algorithms are something you want to know more about, you owe it to yourself to check this out.
Recommended.
Recommended.
Primordial Life --
There's really not a lot of
GA stuff in this game but there is some,
and so I put it here. Besides, it's just plain neat to watch.
Basically, you've got these little A-Life biots that move around the map and search out food. Often that's each other. Where the genetic algorithms come in is in the adaptation of each life form to its environment...neat stuff, and well done.
One rather unique thing about this software is that you can hop on the Internet and connect to other users of the software, allowing your biots to travel across the Internet and battle for survival in other worlds. Neat stuff.
This is just plain fun to watch, by the way. Highly recommended.
Unreal 'Bots --
The use of GP in Unreal 'bots is the
focus of several papers and projects on this fine page by researcher Jorn
Holm. For both his Master's thesis and 7th semester projects he used
genetic programming to build 'bots that
learned and adapted to their environment. Quite interesting stuff.
Full writeups of the theory behind the construction of each 'bot are on the page, as well as sourcecode for a variety of GA and GP projects. Well worth a look, particularly if you're trying to figure out how to use this technology for your own FPS game.
Visual Basic GA Library --
Now this is a little different, a
GA library that's a.) 100% freeware and
b.) written in Visual Basic. You can use it to do all kinds of things
such as solving equations, and it's a great way to investigate how one
might incorporate
GAs into games. You can easily define your
own genome, chromosome length, equations, etc. The source code is fully
commented and avoids usage of the more complicated features of Visual Basic.
The site also contains a very basic HTML tutorial on
GAs that the new user might find useful.