Tell me more ×
Game Development Stack Exchange is a question and answer site for professional and independent game developers. It's 100% free, no registration required.

I am the lead of a research team at my university we are trying to research new developments for artificial intelligence in video games. We are starting with the genre Real-Time-Strategy. The computer science department at my university is really small so I am the only person on the team and i am having difficulty getting people interested in the subject or participating. My team is currently a population of 2.

I have an idea or approach to how i want things to go when we get things rolling but i want to hear what people in game development think i should look at or concentrate on. Do you guys know any interesting games to study? What are popular traits done in industry?

The games i plan to study are the ones i am familair with or can get a huge access of knowledge from. Those being, starcraft 1, starcraft 2, warcraft 3 and age of empires.

This is a very long term project, i just finished my freshman year so i still have alot of learning to do. So for now i am simply doing my classwork and brainstorming until i learn enough to really start doing some stuff. I plan to create general formulas that resemble traits of players based off all the games i decided to study then use those forumulas in my AI.

My objective for the AI is a difficult one something everyone wants basically, i want to make an AI that learns and adapts on its own, uses knowledge from all the games it plays and builds to its arsenal of strategies. Something that is competitive and adaptable according to the player playing it. Also an opponent that will come up with something new on a regular basis but still remain competitive.

As everyone knows, objectives are always subject to change regarding what we can get working and how well everything operates. When i we are ready i am going to enter my AI into the starcraft ai competition using the research that i learned. Also it is probably going to be my senior capstone too. The chair of my department already agreed it would make a good capstone. But that too is subject to change.

My last question. Where are good places to grow interest in something like this and get people participating? I made a forum that i hope people will go to and give feedback/discuss. I am going to operate all my research through that forum (being careful of certain things and not letting people steal information) but for now its not going to be anything special so i don't need to worry about people wanting to steal information XD.

Anyway thanks for your time. I hope i found the right place to post this.

share|improve this question
Warcraft 3 has campaign and multiplayer, in the first the AI doesn't matter because its mostly scripted events and in the second most would play against other humans not against the AI (the skirmish AI of Warcraft 3 is mostly just cheating)... not much AI there. – Mr. Beast Jul 26 '12 at 23:07
1  
What is your research question? You've explained what you want in a few paragraphs but condensing it to one sentence will help you (and us) a lot because it forces you to choose a direction. – Roy T. Jul 27 '12 at 6:44
I want to be able to eventually produce a universal AI that learns and can be implemented in any RTS game enviroment, with as few modifications as possible. – Patrick Murphy Aug 2 '12 at 0:31

Know someone who can answer? Share a link to this question via email, Google+, Twitter, or Facebook.

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.