Uporaba evolucijskih algoritmov pri razvoju igre labirint

magistrsko delo

Tadej Lipar (Author), Iztok Fister (Mentor), Lucija Brezočnik (Co-mentor)

Abstract

Iskanje rešitve za kompleksne probleme je zapletena in zahtevna naloga. Večja, kot je kompleksnost, dlje časa je potrebno vlagati, da pridemo do rešitve. Za hitrejše iskanje rešitev lahko uporabimo evolucijske algoritme, ki temeljijo na Darwinovi evolucijski teoriji. V nalogi smo opisali teorijo evolucijskih algoritmov in podrobneje predstavili genetske algoritme. S pomočjo slednjih smo razvili program, ki generira preproste in rešljive labirinte. Pri tem smo uporabili dva različna načina ocenjevanja kandidatnih rešitev in dobljene rezultate podrobno proučili. Eksperimenti so pokazali, da velikost labirinta močno vpliva na časovno zahtevnost generiranja, da je elitizem bolje ocenjen pristop ter da velikost labirinta in število posameznikov na generacijo pozitivno vplivata na oceno.

Keywords

Evolucijski algoritmi;labirint;genetski algoritmi;igralni pogon Unity;magistrske naloge;

Data

Language:	Slovenian
Year of publishing:	2024
Typology:	2.09 - Master's Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[T. Lipar]
UDC:	004.8.021:004.96(043.2)
COBISS:	202568451
Views:	132
Downloads:	23
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	The use of evolutionary algorithms in the development of a maze game
Secondary abstract:	Finding solutions to complex problems is a complicated and demanding task. The greater the complexity, the more time is required to obtain a solution. Evolutionary algorithms, based on Darwin's evolutionary theory, can be employed to expedite solution-finding. In this thesis, we have described the theory of evolutionary algorithms and have provided a detailed presentation of genetic algorithms. With the help of the latter, we have developed a program that generates simple and solvable mazes. We employed two methods for evaluating candidate solutions and thoroughly examined the results. Experiments have shown that the size of the maze strongly influences the time complexity of generation, that elitism is better evaluated approach, and that both the size of the maze and the number of individuals per generation positively affect the evaluation.
Secondary keywords:	Evolutionary Algorithms;labyrinth;maze;genetic algorithm;Unity game engine;
Type (COBISS):	Master's thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Informatika in tehnologije komuniciranja
Pages:	1 spletni vir (1 datoteka PDF (X, 45 f.))
ID:	23632511