hi.

I'm working on a paper that uses humen as ants to teach bots to choose paths like players. I use ACO to perform this as players move through paths add fromone to path some it adds probability of choosing same path for bots. but not exactly the same path. the point is to make bots path finding adaptive like players and not to find best path. 

the problem is I cant find any work like this for benchmark of my work. can you introduce a work like this to distinguish? thank you for helping.