• norsk
    • English
  • English 
    • norsk
    • English
  • Login
View Item 
  •   Home
  • Fakultet for informasjonsteknologi og elektroteknikk (IE)
  • Institutt for datateknologi og informatikk
  • View Item
  •   Home
  • Fakultet for informasjonsteknologi og elektroteknikk (IE)
  • Institutt for datateknologi og informatikk
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.

Evolving Behaviour Trees: - Automatic Generation of AI Opponents for Real-Time Strategy Games

Hoff, Jonatan Wilhelm; Christensen, Hallvard Jore
Master thesis
Thumbnail
View/Open
12364_FULLTEXT.pdf (2.052Mb)
12364_COVER.pdf (1.556Mb)
URI
http://hdl.handle.net/11250/2405140
Date
2016
Metadata
Show full item record
Collections
  • Institutt for datateknologi og informatikk [3872]
Abstract
Video games are a source of fun and enjoyment for millions of people across the globe. Artificial Intelligence (AI) is an essential part of many games and there is an increasing demand for ever more realistic computer controlled players. There are many methods and approaches for creating AI for games, with state machines and scripting featured in the majority of projects. Behaviour trees have emerged as a recent competitor, combining features of final-state machines and hierarchical task networks. As the available computational powers increase, the feasibility of using evolutionary computations in the development of game AI rises.

This project explores evolving behaviour trees using bio-inspired methods. This is done by tailoring genetic programming to represent individuals as behaviour trees which control a bot that plays a real-time strategy game. The individuals were evaluated by having them compete against each other, a hand-written behaviour tree and an AI bot implemented using traditional methods. Five experiments were conducted using a variety of parameters in order to explore the suitability of using these techniques conjointly.

The results from this project demonstrate that evolving behaviour trees is an interesting technique for automatically generating AI players which can consistently beat ones produced by humans using the same components, although evolving solutions that are serious competitors of traditional AI bots proved more difficult.
Publisher
NTNU

Contact Us | Send Feedback

Privacy policy
DSpace software copyright © 2002-2019  DuraSpace

Service from  Unit
 

 

Browse

ArchiveCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsDocument TypesJournalsThis CollectionBy Issue DateAuthorsTitlesSubjectsDocument TypesJournals

My Account

Login

Statistics

View Usage Statistics

Contact Us | Send Feedback

Privacy policy
DSpace software copyright © 2002-2019  DuraSpace

Service from  Unit