Tehisintellekti loomine käigupõhisele strateegiamängule L-put
Date
2017
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Käigupõhised strateegiamängud on sellised strateegiamängud, milles mängijad langetavad otsuseid käikude kaupa. Need mängud pakuvad suurepärast võimalust rakendada erinevaid tehisintellektide jaoks mõeldud algoritme. Käesoleva töö eesmärk on realiseerida sellist tüüpi mängule tehisintellekt, kasutades minimax ja Monte Carlo algoritme. Töös antakse hinnang nende algoritmide efektiivsusele ning katsetatakse nende võimekust töö raames loodud käigupõhisel strateegiamängul L-put. Samuti annab töö ülevaate kuusnurksete väljadega mänguväljade võimalikest koordinaatsüsteemidest ja algoritmidest, mida nendel mänguväljadel kasutada saab.
Turn-based strategy games are a subset of strategy games, where players make decisions by turns. These type of game provide an excellent opportuniy for testing different Artificial Intelligence algorithms. The aim of this thesis is to create AI systems using minimax and Monte Carlo algorithms and to test their performance on L-put, which is a turnbased strategy game that was made for this thesis. The thesis also provides an overview of different coordinate systems and algorithms for hex-based playing fields.
Turn-based strategy games are a subset of strategy games, where players make decisions by turns. These type of game provide an excellent opportuniy for testing different Artificial Intelligence algorithms. The aim of this thesis is to create AI systems using minimax and Monte Carlo algorithms and to test their performance on L-put, which is a turnbased strategy game that was made for this thesis. The thesis also provides an overview of different coordinate systems and algorithms for hex-based playing fields.