Procedural urban environments for FPS games

Jan Kruse; Ricardo Sosa; Andy M. Connor

doi:10.1145/2843043.2843479

Procedural urban environments for FPS games

Jan Kruse, Ricardo Sosa, Andy M. Connor

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

Abstract

This paper presents a novel approach to procedural generation of urban maps for First Person Shooter (FPS) games. A multi-agent evolutionary system is employed to place streets, buildings and other items inside the Unity3D game engine, resulting in playable video game levels. A computational agent is trained using machine learning techniques to capture the intent of the game designer as part of the multi-agent system, and to enable a semi-automated aesthetic selection for the underlying genetic algorithm.

Original language	English
Title of host publication	Interactive Entertainment 2016, 2-5 February, Canberra, Australia
Subtitle of host publication	[part of the] Australasian Computer Science Week (ACSW 2016) [proceedings]
Place of Publication	New York, NY
Publisher	Association for Computing Machinery (ACM)
Number of pages	5
ISBN (Electronic)	9781450340427
DOIs	https://doi.org/10.1145/2843043.2843479
Publication status	Published - 1 Feb 2016
Externally published	Yes
Event	Interactive Entertainment 2016 - Canberra, Australia Duration: 2 Feb 2016 → 5 Feb 2016 http://ieconference.org/ie2016/

Conference

Conference	Interactive Entertainment 2016
Abbreviated title	IE 2016
Country	Australia
City	Canberra
Period	2/02/16 → 5/02/16
Internet address	http://ieconference.org/ie2016/

Keywords

Agents
Artificial intelligence
Computer games
First person shooter, urban environment
Genetic algorithm
Procedural environment
Unity

Access to Document

10.1145/2843043.2843479

Link to publication in Scopus

Cite this

@inproceedings{fad6c931d6924bc595dfacee422e0059,

title = "Procedural urban environments for FPS games",

abstract = "This paper presents a novel approach to procedural generation of urban maps for First Person Shooter (FPS) games. A multi-agent evolutionary system is employed to place streets, buildings and other items inside the Unity3D game engine, resulting in playable video game levels. A computational agent is trained using machine learning techniques to capture the intent of the game designer as part of the multi-agent system, and to enable a semi-automated aesthetic selection for the underlying genetic algorithm.",

keywords = "Agents, Artificial intelligence, Computer games, First person shooter, urban environment, Genetic algorithm, Procedural environment, Unity",

author = "Jan Kruse and Ricardo Sosa and Connor, {Andy M.}",

year = "2016",

month = feb,

day = "1",

doi = "10.1145/2843043.2843479",

language = "English",

booktitle = "Interactive Entertainment 2016, 2-5 February, Canberra, Australia",

publisher = "Association for Computing Machinery (ACM)",

address = "United States of America",

note = "Interactive Entertainment 2016, IE 2016 ; Conference date: 02-02-2016 Through 05-02-2016",

url = "http://ieconference.org/ie2016/",

}

Kruse, J, Sosa, R & Connor, AM 2016, Procedural urban environments for FPS games. in Interactive Entertainment 2016, 2-5 February, Canberra, Australia: [part of the] Australasian Computer Science Week (ACSW 2016) [proceedings]., a77, Association for Computing Machinery (ACM), New York, NY, Interactive Entertainment 2016, Canberra, Australia, 2/02/16. https://doi.org/10.1145/2843043.2843479

Procedural urban environments for FPS games. / Kruse, Jan; Sosa, Ricardo; Connor, Andy M.

Interactive Entertainment 2016, 2-5 February, Canberra, Australia: [part of the] Australasian Computer Science Week (ACSW 2016) [proceedings]. New York, NY : Association for Computing Machinery (ACM), 2016. a77.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Procedural urban environments for FPS games

AU - Kruse, Jan

AU - Sosa, Ricardo

AU - Connor, Andy M.

PY - 2016/2/1

Y1 - 2016/2/1

N2 - This paper presents a novel approach to procedural generation of urban maps for First Person Shooter (FPS) games. A multi-agent evolutionary system is employed to place streets, buildings and other items inside the Unity3D game engine, resulting in playable video game levels. A computational agent is trained using machine learning techniques to capture the intent of the game designer as part of the multi-agent system, and to enable a semi-automated aesthetic selection for the underlying genetic algorithm.

AB - This paper presents a novel approach to procedural generation of urban maps for First Person Shooter (FPS) games. A multi-agent evolutionary system is employed to place streets, buildings and other items inside the Unity3D game engine, resulting in playable video game levels. A computational agent is trained using machine learning techniques to capture the intent of the game designer as part of the multi-agent system, and to enable a semi-automated aesthetic selection for the underlying genetic algorithm.

KW - Agents

KW - Artificial intelligence

KW - Computer games

KW - First person shooter, urban environment

KW - Genetic algorithm

KW - Procedural environment

KW - Unity

UR - http://www.scopus.com/inward/record.url?scp=84962523486&partnerID=8YFLogxK

U2 - 10.1145/2843043.2843479

DO - 10.1145/2843043.2843479

M3 - Conference Paper

AN - SCOPUS:84962523486

BT - Interactive Entertainment 2016, 2-5 February, Canberra, Australia

PB - Association for Computing Machinery (ACM)

CY - New York, NY

T2 - Interactive Entertainment 2016

Y2 - 2 February 2016 through 5 February 2016

ER -