Evolutionary programming for automatic generation of military strategies in RTS games

Huat, Ch'ng Siong and Teo, Jason Tze Wi and Norhayati Daut (2008) Evolutionary programming for automatic generation of military strategies in RTS games. In: 13th International Conference on Computer Games (CGames 2008) , 03-05 November 2008 , Wolverhampton, England.

Full text not available from this repository.


Evolutionary Algorithms (EAs) are commonly used for finding optimal solutions in large and difficult search spaces. In real-time strategy (RTS) games, huge search spaces in addition to the highly dynamic environments constitute tremendous challenges for the generation of game artificial intelligence (AI), yet there have been very few attempts to date that use EAs to optimize game AI in RTS games. This paper describes an experiment regarding how RTS military strategies can be optimized using a branch of EAs known as Evolutionary Programming (EP). To reduce the effect of dynamic noise arising from the highly stochastic RTS playing environment, a (mu+mu) EP survivor selection strategy is applied. The results from the experiment clearly show that firstly, the problem is non-trivial and more importantly, the (mu+mu) EP methodology is able to automatically generate good military strategies and significantly outperforms the control setup of using randomly generated strategies. Moreover, the results also show that the evolutionary optimization approach could learn very fast and can thus be practically used as an online game AI evolutionary method against human players, which can then make the game more challenging and interesting for the human players.

Item Type: Conference or Workshop Item (UNSPECIFIED)
Keyword: Military strategies, RTS, Real-time strategy games, Computer science, Evolutionary algorithms
Subjects: Q Science > QA Mathematics > QA1-939 Mathematics > QA71-90 Instruments and machines > QA75.5-76.95 Electronic computers. Computer science
Department: SCHOOL > School of Engineering and Information Technology
Depositing User: ADMIN ADMIN
Date Deposited: 09 Nov 2011 16:05
Last Modified: 30 Dec 2014 14:47
URI: https://eprints.ums.edu.my/id/eprint/1241

Actions (login required)

View Item View Item