Application of genetic algorithm in games

OData support
Dr. Horváth Gábor
Department of Measurement and Information Systems

In most different types of games there is a need of doing some kinds of optimization. The optimization can be needed for solving a subproblem, but winnig the game can be a optimum-finding search. Several algorithmic solutions have been developed to these over the years. I tried to find out these algorithms and to show their evolution in the game industry.

In my thesis I examined the genetic algorithms’ structure and behaviour. I considered how it can be built in the world of games and what kind of subproblems can be solved using these methods. Where I could, I brought examples of their usage to illustrate the wide range of options.

The other purpose of my thesis was to create a Tower Defense game. I wrote the game’s engine and graphical interface, too. Then, I thought about which point I can use genetic algorithms and I accomplished one of them. My goal was that the player should feel that the computer reacts to his or her movements during the game. For that, I performed experiments with different parameters of the genetic algorithms.


Please sign in to download the files of this thesis.