Artificial Intelligence for Chess Engine

OData support
Rajacsics Tamás
Department of Automation and Applied Informatics

Since the dawn of computer science informaticians are interested in creating

artificial intelligence. The first stepping stone of this process was the creation of a

computer that can play chess. Since the fifties, programmers try to create more and

more sophisticated algorithms in order to overcome human chess players. This goal was

finally achieved in 1997 when Deep Blue defeated the world champion Garry Kasparov

in their match. Nowadays a stronger chess engine running on a laptop or personal

computer is able to defeat the strongest grandmasters of the world. This is the result on

one hand of the increasing computational power of the computers, on the other hand of

the faster and better algorithms worked out by the experts of the field.

Modern chess engines can be separated to two main parts. The board

representation is responsible for the rules of the game and the position of the pieces and

the artificial intelligence for the decision making process. My task was to plan and

implement the decision making module of the engine in a way, that it can work

together with a board representation and play successfully. The other module was

implemented by Boros Dávid in his thesis called „Board representation and framework

for chess engine”.

In the first part of the thesis I present the basic structure of the chess engine, it’s

inner workings with the main emphasis on the decision making. I describe the different

methods and implement them. At the end of the paper I test and optimize the produced

search and heuristic algorithms.


Please sign in to download the files of this thesis.