Skip to main navigation Skip to search Skip to main content

Evolving believable behaviours for first person shooter games

  • University of Galway

Research output: Chapter in Book or Conference Publication/ProceedingConference Publicationpeer-review

Abstract

This papers presents research in the use of genetic programming techniques to generate believable behaviours for NPCs in first person shooter games. We use the Quake II game and engine in this work to illustrate our approach. We report aspects of the NPC behaviours deemed unrealistic by a number of human game players. We abstract these scenarios into an abstract game wherein we evolve fit behaviours for our NPCs. These evolved scripts are then placed back into the original Quake II engine. We report on the evolved strategies in the abstract game and also on feedback from a survey of game players on their view of these behaviours in the game engine.

Original languageEnglish
Title of host publication9th International Conference on Intelligent Games and Simulation, GAME-ON 2008
Pages44-48
Number of pages5
Publication statusPublished - 2008
Event9th International Conference on Intelligent Games and Simulation, GAME-ON 2008 - Valencia, Spain
Duration: 17 Nov 200819 Nov 2008

Publication series

Name9th International Conference on Intelligent Games and Simulation, GAME-ON 2008

Conference

Conference9th International Conference on Intelligent Games and Simulation, GAME-ON 2008
Country/TerritorySpain
CityValencia
Period17/11/0819/11/08

Keywords

  • Artificial intelligence

Fingerprint

Dive into the research topics of 'Evolving believable behaviours for first person shooter games'. Together they form a unique fingerprint.

Cite this