Artificial Intelligence Depot
Visiting guest. Why not sign in?
News, knowledge and discussion for the AI enthusiast.
FEATURES COMMUNITY KNOWLEDGE SEARCH  
Get Creative: AI Article Writing Contest
Fancy the chance of getting developer focus, improving your research skills, sharing your artificial intelligence ideas, obtaining expert feedback, getting published online AND winning a prize?
Enter the AI Article Writing Contest!
Academia AI in Game Developing
using A-Life knowledge to make better games
 
Academia AI in Game Developing

Our goal is to make better games. We (AI Lab Zurich, Switzerland) have recently established a group (all people with stupendous gaming background :)) within our lab which is currently studying communication possibilities to the commercial game developing industry, concerning possible partnerships and networking with companies. We have some researchers who are working on virtual reality modelling in the field of artificial life (Artificial Ontogeny, GA) and who are more than willing to share their know-how. We believe that applied A.I. (the way WE look at it) could be a highly powerful (and refreshing) component to game developing. We also have some concrete and realistic ideas on how our knowledge could be implemented in this field (specially AI-Life: morphogenetic design / evolutionary algorithms):

We have developed a tool, "MorphEngine", that allows student with little prior knowledge to evolve fantastic artificial creatures. This has entertainment value, but can also be used to communicate highly difficult academic materials about evolution.

One of the major advantages of “morphogenetic design” is that we can minimize the designer biases in the design process. As long as one has to parameterize the final structure, one has to have a pretty good idea about this structure, thus, the designer biases tend to be very strong. Minimizing these biases promises novel solutions.

Nevertheless, in order to achieve some kind of actual link to the game developping community, we are in need of help. Since we're a relatively "young" group at our lab, we are looking for people in the community who are willing to point us in the right direction, to support our actions or to work with us in some sort way. If you, by any chance, are interested in what we do and if you see any possibility of creating an exchange of thoughts in this matter, or even if you know who specifically might be interested in our request, I'd be more than happy to give you concrete information on our research field. Feel free to visit our website at: www.ifi.unizh.ch/ailab.

Greetings to the AI community,

R.S.

2 posts.
Monday 06 January, 05:07
Reply
GDC academic summit / connecting to game developer problems

I guess you already intend to visit the Academic Summit on the upcoming GDC (first week of March, San Jose, Ca.). Otherwise, you definitely should consider attending:
http://www.gdconf.com/conference/academicsummit.htm

Please note that your statement:
"We believe that applied A.I. (the way WE look at it) could be a highly powerful (and refreshing) component to game developing. We also have some concrete and realistic ideas on how our knowledge could be implemented in this field (specially AI-Life: morphogenetic design / evolutionary algorithms):"

is a nice example of "technology push".

You will have a hard time convincing game developers, unless you are first able to:
- identify the top 3 "ai related" problems a game developer today has;
- how your technology tackles one of these problems without causing big problems for other parts of the game code, content, content pipe line, play balancing, planning, and training.

Doug Church does an excellent job at illustrating the role of AI in game development:
http://ai.eecs.umich.edu/people/laird/game-seminar/Doug-Church.ppt

William

19 posts.
Thursday 16 January, 05:27
Reply
top 3 "AI related" problems

Hey William, thank you for your reply, it's very much appreciated! I've
been looking into that Academic Summit Website you mentioned, very
interesting. Unfortunately though, it's very soon so our institute won't be able to make it.
Now concerning your message, I'm not quite sure I understand what you
mean by: "identify the top 3 'ai related' problems a game developer
today has". Could you be more a little more specific? We are aware that
implementing A.I. into a game is not a trivial thing, but there are
different categories like for example the structure of a crowd (i.e. at
a soccer stadium), flocking behaviour on creatures, intelligent
behaviour of bots, evolving creatures or even the fluent design of
environment/background.

We are interested in creating holistic creatures mainly. Agents that
move in a very realistic way, depending on the physical laws of the
environment of interest. One example of approach: We achieve this by
giving it a task (swim, fly, push, run,...) and physical laws and
evolving them for many generations. In addition, we take the evolved
"exoskeleton", then use a special program to make them move smoothly and realistic (the texture will also be evolved generatevely). A creature foundry, if you will. We soon will have our website up and I'd be very happy to update you on our project.

Best,

Rafael

2 posts.
Monday 27 January, 06:52
Reply