Artificial Intelligence for the OpenTTD Game

Technical University of Denmark

Department of Informatics and Mathematical Modeling

Master Thesis

Artificial Intelligence for the OpenTTD Game

Author: Maciej Wisniewski

Supervisor: Dr. Carsten Witt

Kongens Lyngby 2011 IMM-M.Sc.-2011-56

Technical University of Denmark Informatics and Mathematical Modelling Building 321, DK-2800 Kongens Lyngby, Denmark Phone +45 45253351, Fax +45 45882673 reception@imm.dtu.dk imm.dtu.dk

IMM-M.Sc.: ISSN 0909-3192

Abstract

This master thesis project report is a result of an analysis of artificial intelligence applications in the field of transport management, focusing on optimal economic strategy and based on the example of OpenTTD, a simulation game. During this analysis a custom artificial intelligent agent has been designed and implemented for this game. OpenTTD is a simulation game available for free on its website, as an open-source project. The objective of the game is to create and manage your own transport company and potentially achieve the best performance company ratings. The project presented had several specified aims. In the beginning the author gives a brief game description and describes related problems which are solvable through the usage of knowledge from the field of artificial intelligence. In the next chapters an analysis of human behaviors, playing strategies and a typical human approach to the game is presented. The game API and artificial intelligence implementation language Squirrel are investigated and learned. Based on the gained knowledge existing artificial intelligence implementations and their designs are analyzed and described. Comparison classes and properties for categorization of artificial players are determined. Using these results and the gathered knowledge an artificial intelligence called SPRING is implemented and presented. Afterwards, an evaluation of the results based on selected test cases is made. In the end of the report a discussion (with examples) of real-life applications is presented. Additionally, a short discussion of possible game improvements in the area of artificial intelligence is presented.

iii

Acknowledgements

I would like to thank my supervisor for a lot of patience and good advice. Likewise, I would like to thank to my family and friends for the support they have provided.

Maciej Wisniewski

v

 ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download