Complex Data Structures with Genetic Algorithms: Can a Computer make Exciting Rollercoasters in Rollercoaster Tycoon 2?

  • Joshua William Leland Department of Electronics, Computing, & Mathematics

Abstract

The field of Procedural Content Generation (PCG) has been the topic of much research in recent years, their application to games specifically has seen a huge increase in the number of studies using some form of generation to produce game content. In particular, Genetic Algorithms (GAs) and other forms of machine learning have seen use in the creation of bespoke content for particular games. This paper covers the creation of a GA, dubbed RCT2GA, that is tasked with the creation of creative, exciting, and believable content in the context of the 2002 game Rollercoaster Tycoon 2. RCT2GA is based upon methodologies proposed by Kevin Burke, and seeks to improve upon the groundwork set. Additionally, multiple problems are discovered throughout development, with workarounds, solutions, and key areas for future study being suggested as a result.
Published
2017-06-17
How to Cite
LELAND, Joshua William. Complex Data Structures with Genetic Algorithms: Can a Computer make Exciting Rollercoasters in Rollercoaster Tycoon 2?. Discovery, Invention & Application, [S.l.], june 2017. Available at: <https://computing.derby.ac.uk/ojs/index.php/da/article/view/259>. Date accessed: 25 feb. 2018.
Issue
Section
Articles