Evolving Battle Formations in Massively Multiplayer Online Strategy Games

This paper presents a genetic algorithm for the development of battle formations for MMORTS games. We considered the context of the game Call of Roma, where the battles are turn based, where only two sides fight each other, which are the Attack and the Defense. The genetic algorithm takes as input a predetermined battle formation, presented by the defense, and returns a battle formation adapted for the Attack. The algorithm aims to maximize the balance of the Attack side. This balance is formed by the difference between the number of slaughtered soldiers in combat on both sides. The balance is used to calculate the fitness of individuals, which encode the various characteristics of the fighting heroes. The results demonstrate that the proposed algorithm is able to find a victorious solution for the Attack, even when it is under unfavorable conditions.