Autonomna vožnja korištenjem dubokog potpornog učenja

undergraduate thesis

• Title (english): Autonomous driving using deep reinforcement learning
• Author: Leon Bertol
• Mentor: Marija Seder (mentor)
• Committee member: Marija Seder (predsjednik povjerenstva)
• Committee member: Filip Marić (član povjerenstva)
• Committee member: Borna Bićanić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-02, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Autonomna vožnja jedan je od problema kojim se bavi umjetna inteligencija. Kako bi vozilo samostalno upravljalo potrebna je velika količina podataka koja se postiže ugradnjom različitih vrsta senzora. Za obradu i učenje stečenih informacija koriste se tehnike strojnog učenja, a u tu svrhu, duboko podržano učenje svojim nedavnim razvojem je postalo izvrstan kandidat . U ovom radu provodi se primjena i oblikovanje problema dubokog potpornog učenja u autonomnoj vožnji te su dobiveni rezultati prezentirani i interpretirani. Također kako bi se naznačilo koji senzori daju najbolje rezultate, testirana su tri modela koja koriste različite vrste senzora.

Abstract (english)

Autonomous driving is one of the problem that is dealt by artificial intelligence. To drive autonomously vehicle requires vast quantity of information which is achieved by installing various types of sensors. For processing and learning from information machine learning techniques are used and recent development of deep reinforcement learning has made it superb candidate for that purpose. In this paper such application and shaping of deep reinforcement learning problem on autonomous driving is implemented and acquired results are presented and interpreted. Also to indicate which combination of observation gives the best results three models using different observations were tested.

• Keywords: duboko podržano učenje; autonomna vožnja; duboke neuronske mreže; duboko učenje; strojno učenje; CARLA simulator; Soft Actor Critic
• Keywords (english): deep reinforcement learning; autonomous driving; deep neural networks; deep learning; machine learning; CARLA simulator; Soft Actor Critic
• Language: english
• URN:NBN: urn:nbn:hr:168:915767
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 20:05:20