master's thesis
Tim Štromajer (Author), Iztok Lebar Bajec (Mentor), Jure Demšar (Co-mentor), Afshin Ameri (Co-mentor)

Abstract

Different organisms tend to form spontaneous and less predictable groups of individuals while doing everyday activities, such as eating or migrating. By understanding the rules of so called collective behaviour, we can learn how to control these groups to one's desires. Similar phenomena is of interest in many other domains like crowd control, cleaning the environment and other engineering problems. In this work, we focus on shepherding, which is an act of influencing or herding a flock of sheep using a shepherd dog. Here we create a model based on the existing shepherd dog models and then present some improvements to it to make it more realistic, such as limit the vision, add ability to hear other animals and implement a short term memory. We also adapt the model to allow multiple shepherd dogs to herd sheep at the same time. After that we present a model that is not based on some predefined rules, but is trained using reinforcement learning. Four different dog models are created, each able to observe the environment in a different way. The results show that the best model is the one that is using a ray casting method for observation.

Keywords

shepherd dogs;collective behaviour;shepherding;agent models;artificial intelligence;reinforcement learning;simulation;computer science;master's thesis;

Data

Language: English
Year of publishing:
Typology: 2.09 - Master's Thesis
Organization: UL FRI - Faculty of Computer and Information Science
Publisher: [T. Štromajer]
UDC: 004.8:636.7.043.7(043.2)
COBISS: 124498691 Link will open in a new window
Views: 26
Downloads: 8
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: Slovenian
Secondary title: Uporaba strojnega učenja za učenje psa ovčarja
Secondary abstract: Številni organizmi velikokrat oblikujejo spontane in manj predvidljive skupine posameznikov, medtem ko opravljajo vsakodnevne dejavnosti, kot so prehranjevanje ali selitev. Z razumevanjem pravil tako imenovanega kolektivnega vedenja, se lahko naučimo, kako te skupine nadzorovati po svojih željah. S podobno tematiko se soočajo tudi na številnih drugih področjih, kot so nadzor množice, čiščenje okolja in drugi inženirski problemi. V tem delu smo se osredotočili na pastirstvo, kjer psi ovčarji s pomočjo strahu pasejo in preganjajo čredo ovc. Naredili smo model, ki temelji na obstoječih modelih psov ovčarjev in ovc, ter mu dodali nekaj izboljšav, da postane bolj realističen, kot so omejitev vida, sposobnost sluha in implementacija kratkoročnega spomina. Model smo prilagodili tudi tako, da omogoča sodelovanje večih psov pri pašnji črede. V nalogi predstavimo tudi modele, ki ne temeljio na vnaprej določenih pravilih, ampak se psi naučijo pasti ovce s pomočjo spodbujevanega učenja. Ustvarimo štiri takšne modele psov ovčarjev, od katerih vsak pridobiva podatke iz okolja na drugačen način. Z rezultati pokažemo, da se za najboljši model izkaže tisti, ki za opazovanje uporablja metodo sledenja žarkov.
Secondary keywords: psi ovčarji;pastirstvo;agentni modeli;umetna inteligenca;spodbujevano učenje;simulacija;magisteriji;Strojno učenje;Kolektivno vedenje;Psi;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: VI, 57 str.
ID: 16653088