Automated Game Balancing in Ms PacMan and StarCraft Using Evolutionary Algorithms

Conference paper

DOI: 10.1007/978-3-319-55849-3_25

Part of the Lecture Notes in Computer Science book series (LNCS, volume 10199)
Cite this paper as:
Morosan M., Poli R. (2017) Automated Game Balancing in Ms PacMan and StarCraft Using Evolutionary Algorithms. In: Squillero G., Sim K. (eds) Applications of Evolutionary Computation. EvoApplications 2017. Lecture Notes in Computer Science, vol 10199. Springer, Cham

Abstract

Games, particularly online games, have an ongoing requirement to exhibit the ability to react to player behaviour and change their mechanics and available tools to keep their audience both entertained and feeling that their strategic choices and in-game decisions have value. Game designers invest time both gathering data and analysing it to introduce minor changes that bring their game closer to a state of balance, a task with a lot of potential that has recently come to the attention of researchers. This paper first provides a method for automating the process of finding the best game parameters to reduce the difficulty of Ms PacMan through the use of evolutionary algorithms and then applies the same method to a much more complex and commercially successful PC game, StarCraft, to curb the prowess of a dominant strategy. Results show both significant promise and several avenues for future improvement that may lead to a useful balancing tool for the games industry.

Keywords

Evolutionary algorithms Game balance Automation PacMan StarCraft 

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.University of EssexColchesterUK

Personalised recommendations