Gerrits, J.G. (2009) Optimizing a team of quake 3 bots for the capture the flag game type. Master's Thesis / Essay, Artificial Intelligence.
|
Text
AI-MAI-2009-J.G.GERRITS.pdf - Published Version Download (1MB) | Preview |
Abstract
In recent years we have seen a huge increase in the popularity of video games. The market for video games has already outgrown the movie industry with annual sales over 31 billion dollars. With the decreasing costs of computational power, the graphics and physics of the games become more and more realistic. However, these are not the only features that make a game successful. The behavior of non-playing characters (NPC's) is equally important to make a game realistic, and thereby fun to play. The industry does not seem to be able to tackle this problem. Even for recently released games, the community still is not satisfied with the game AI at all. In this research, a genetic algorithm and a particle swarm optimizer will be used to train the neural networks of a team of NPC's for the 'capture the flag' game type in Quake III Arena. The teams will be trained with and without a means of communication. Results show that all methods perform better than the fuzzy logic approach of the Quake III engine.
| Item Type: | Thesis (Master's Thesis / Essay) |
|---|---|
| Degree programme: | Artificial Intelligence |
| Thesis type: | Master's Thesis / Essay |
| Language: | English |
| Date Deposited: | 15 Feb 2018 07:28 |
| Last Modified: | 15 Feb 2018 07:28 |
| URI: | https://fse.studenttheses.ub.rug.nl/id/eprint/8581 |
Actions (login required)
 |
View Item |