magistrsko delo
Simon Dimc (Author), Iztok Lebar Bajec (Mentor), Jure Demšar (Co-mentor)

Abstract

Metode mehke logike omogočajo definiranje skupinskega vedenja množice agentov. Definiranje pravil z metodami mehke logike je vizualno preprosto, izračunano vedenja pa precej naravno. V filmski industriji se že nekaj časa uporabljajo orodja, ki ponujajo takšne metode definiranja vedenja množic. Pri razvoju iger pa v javno dostopnih igralnih okoljih teh še ne najdemo. V magistrskem delu za igralno okolje Unity razvijemo ogrodje za vizualno programiranje vedenja množice agentov. Sestavljeno je iz treh večjih komponent: urejevalnika vizualnega programiranja vedenja, komponente za nastavljanje lastnosti agenta in komponente za upravljanje računanja vedenja. Ogrodje testiramo na treh modelih Boids: letenje jate ptic v 2D in 3D prostoru ter hoja ljudi proti skupnemu cilju, skozi poligon z ovirami. Pri modelih letenja jate ptic izmerimo učinkovitost računanja vedenja. Merimo število slik na sekundo pri različnih nastavitvah računanja vedenja. Velikost množice agentov, pri kateri se simulacija izvaja pri 60 slikah na sekundo, je 300 v 2D prostoru in 200 v 3D prostoru.

Keywords

računalniške igre;skupinsko vedenje;simulacija;računalništvo in informatika;magisteriji;

Data

Language: Slovenian
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [S. Dimc]
UDC: 004.94(043.2)
COBISS: 129727491 Link will open in a new window
Views: 45
Downloads: 10
Average score: 0 (0 votes)
Metadata: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Other data

Secondary language: English
Secondary title: Fuzzy crowd simulations for computer games
Secondary abstract: Fuzzy logic methods can be used for defining collective behaviour of a group of agents. Defining the rules with fuzzy logic is visually simple and the resulting behaviour is quite natural. Tools that offer such methods for defining collective behaviour have been in use in film industry for some time now. But when we look at game development, specifically in public game engines, we don't find them yet. In this master's thesis we develop a framework for Unity game engine that offers visual programming of collective behaviour. It consists of three major components: an editor for visual programming of behaviour, a component for setting the agent properties and a component for managing behaviour computation settings. We test the framework on three Boids models: flight of a flock of birds in 2D and 3D space, and people walking towards a common goal, through a field with obstacles. We measure the efficiency of behaviour computation on flocks of birds models. We measure the frames per second at different behaviour computation settings. The size of the group of agents, at which the simulation runs at 60 frames per second, is 300 in 2D space and 200 in 3D space.
Secondary keywords: computer games;collective behaviour;fuzzy logic;computer science;computer and information science;master's degree;Videoigre;Mehka logika;Simulacijske metode;Računalništvo;Univerzitetna in visokošolska dela;
Type (COBISS): Master's thesis/paper
Study programme: 1000471
Embargo end date (OpenAIRE): 1970-01-01
Thesis comment: Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages: 48 str.
ID: 17034224