Lots of food for thought there. Clearly I didn't have enough math in college.
For global ant allocation and food gathering, it's very possible that the approach you're describing is the best you could practically do. For more local combat situations, though, I don't think that's clear at all-- if I had infinite computing resources, I'd be using an appropriate adaptation of minimax.
Anyway, you gave me some words for concepts and some new concepts, so upvote for you.
Aichallenge.org has started their third AI contest this year: Ants.
I mentioned this in the open thread, and there was a discussion about possibly making one or more "official" LessWrong teams. D_Alex has offered a motivational prize. If this interests you, please discuss in the comments!