Galen, do you have any sources I could read up on?
I've done, connect4(perfect), several other board games, and have done a few competitive robotics competitions (First Robotics). I am currently taking on a 2d space game though and I am kind of lost on how to approach it.
There's a bunch of different AI approaches.
The first is used in games, and is the most easy. It uses simple logic, state machines, and a variety of other primarily deterministic algorithms, with a bit of fuzzy logic mixed in. IMO, the boring crap which is more 'programmer common sense' than anything else.
The second is the connectionism approach. Neural networks and similar. They will probably do what you want most of the time if you do it right, but you probably won't know how or why. Good for pattern recognition.
The third is the evolutionary approach. Genetic algorithms. You pretty much give it a goal, then it tries to get as close to the goal as possible, even if the problem isn't well defined. Good if you know what you want, but not how you want it. Good for occasionally creative solutions.
The most recent and most successful variation is mixing in Bayesian Reasoning. Markov models, bayesian networks, and a few others. Based very heavily in statistics. They can be used on their own or in combination with subsystems consisting of other AI techniques; if you want to learn more about these, go here:
https://www.ai-class.com/home/Seriously, it deserves that size. It's the online course that pretty much started the whole 'coursera' thing, and basically will take you from beginner level AI all the way up to the level where you could probably code a self-driving car. The course part is over, and so there aren't the homeworks or exam problems, but those don't really add anything anyway. It also has comprehension check questions scattered throughout, with explanations if you didn't understand the concept.
The guys teaching it are as good as it gets. Peter Norvig literally wrote the book on AI; "Artificial Intelligence: A Modern Approach," which is pretty much
the book on AI. Sebastian Thrun was the head of the team that created Stanford's self driving car, the first to complete the DARPA Grand Challenge, before moving on to Google to create the Google self driving car. So yeah; basically the top guys in the field.