An Evolutionary Strategy for a Computer Team Game

Abstract

Computer team games have attracted many players in recent years. Most of them are rule-based systems because they are simple and easy to implement. However, they usually cause a game agent to be inflexible, and it may repeat a failure. Some studies investigated the learning of a single game agent, and its learning capability has been improved. However, each agent in a team is independent and it does not cooperate with others in a multiplayer game. This article explores an evolution strategy for a computer team game based on Quake III Arena. The Particle Swarm Optimization (PSO) algorithm will be applied to evolve a non-player character (NPC) team in Quake III to be more efficient and intelligent. The evolution of a single NPC, which accommodates to its team and, moreover, the team has learning and cooperating abilities, will be discussed. An efficient team is composed of various members with their own specialties, and the leader is capable of evaluating the performance of a member and assigning it a proper job. Furthermore, the leader of an intelligent team will adapt a strategy appropriate for various circumstances and obtain the team's best performance. Instead of considering the tactic of an individual bot, this article takes the strategy of a team into account.

Evolving a Team in a First-Person Shooter Game by Using a Genetic Algorithm

“…. , 7g), a l is the weight of the transition, and t l is the transition time between two consecutive states (Tsai et al 2011). The fitness value of a team is evaluated after a match is over, with an optimal strategy having a higher fitness value.…”

“…It is applied to evolve agents in games such as tic-tac-toe, checkers, and others (Messerschmidt and Engelbrecht 2004;Engelbrecht 2005Engelbrecht , 2006Duro and de Oliveira 2008;Tsai et al 2011). PSO has the advantage of being able to converge fast; however, a local optimum is likely acquired instead of obtaining a global optimum.…”

Evolving a Character in a First-person Shooter

“…The Particle Swarm Optimization (PSO) was applied to evolve game agents. (Messerschmidt and Engelbrecht, 2004); (Tsai et al, 2011). PSO has the advantage of being able to converge fast; however, a local optimum is likely acquired instead of obtaining a global optimum (Shyr, 2008).…”