Primjena metoda podržanog učenja u domeni autonomnog parkiranja

master's thesis

• Title (english): Leveraging Reinforced Learning Methods in the Domain of Autonomous Parking
• Author: Marko Marfat
• Mentor: Damir Pintar (mentor)
• Committee member: Damir Pintar (predsjednik povjerenstva)
• Committee member: Mihaela Vranić (član povjerenstva)
• Committee member: Luka Humski (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-13, Croatia

Abstract

Podržano učenje je dio strojnog učenja koji se bavi učenjem modela umjetne
inteligencije kako bi bio računalni sustav bio sposoban autonomno donositi odluke ovisno
o parametrima svojeg okruženja. Model s vremenom nauči maksimizirati nagradu koju
dobije kada donese ispravnu odluku. Zadatak diplomskog rada je razviti model za
autonomno parkiranje koristeći metode podržanog učenja. U radu će se uz pomoć metoda
podržanog učenja oblikovati programska simulacija čija je svrha omogućiti računalu
automatsko učenje upravljanja modela virtualnog vozila kako bi se isto moglo parkirati na
simuliranom parkingu s raznim preprekama. Rad će također istražiti i primjenu dobivenih
rezultata u stvarnom svijetu koristeći manji model vozila. Cilj je pružiti uvid u potencijal
podržanog učenja u domeni autonomne vožnje i u razvoj inteligentnih autonomnih sustava.

Abstract (english)

Reinforced learning is a subset of machine learning that specializes in training AI
models in such a way that a computer system is able to make decisions autonomously
depending on the parameters of its environment. Over time, the model learns to maximize
the reward it gets when it makes the right decision. The goal of the thesis is to develop a
model for autonomous parking using reinforced learning methods. In the thesis, with the
help of reinforced learning methods, a software simulation will be created whose purpose
is to enable the computer to automatically learn to control a virtual vehicle model with the
goal of being able to park in a simulated parking lot with various obstacles. The thesis will
also explore the application of the obtained results in the real world using a small model
vehicle. The goal is to provide insight into the potential of reinforced learning in the
domain of autonomous driving and into the development of intelligent autonomous
systems.

• Keywords: podržano učenje; autonomno parkiranje; neuronske mreže; nadzirano učenje
• Keywords (english): reinforced learning; autonomous parking; neural networks; supervised learning
• Language: croatian
• URN:NBN: urn:nbn:hr:168:480417
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-10-11 09:39:31