Autonomno istraživanje nepoznatih velikih unutarnjih prostora mobilnim robotom za izgradnju trodimenzionalnoga modela

doctoral thesis

• Title (english): Autonomous exploration of unknown large indoor environments by a mobile robot for three-dimensional model building
• Author: Ivan Maurović
• Mentor: Ivan Petrović (mentor)
• Committee member: Ivan Petrović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Control and Computer Engineering); Zagreb
• Defense date and country: 2017, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering; Automation and Robotics
• Universal decimal classification (UDC): 621.3; APPLIED SCIENCES. MEDICINE. TECHNOLOGY; Electrical engineering

Abstract

U ovome doktorskom radu opisano je istraživanje nepoznatog unutarnjeg prostora mobilnim robotom. Zadatak je algoritma istraživanja odrediti minimalan broj točaka s kojih će se snimiti čitavi prostor. U radu je razvijen algoritam kojim se smanjuje složenost proračuna kombinirajući istraživanje prostora zasnovano na 2D podacima i istraživanje zasnovano na 3D podacima sa senzora udaljenosti. Kako bi se omogućilo prebacivanje istraživanja između 2D i 3D pristupa, razvijen je algoritam izdvajanja sobe koji na temelju 2D podataka s laserskog senzora udaljenosti izdvaja sobu. U radu je opisan razvijeni algoritam planiranja putanje gibanja robota kojom robot stiže do cilja izbjegavajući prepreke te vodeći računa o lokalizaciji robota. Razvijeni algoritam temelji se na D* algoritmu planiranja putanje robota koji je u radu prilagođen za pronalazak najkraćeg puta u grafu s negativnim cijena prijelaza između čvorova. Algoritam pronalaska optimalne putanje provjeren je simulacijski i eksperimentalno na mobilnom robotu. U ovoj doktorskoj disertaciji ostvaren je sljedeći znanstveni doprinos: - algoritam za određivanje skupa diskretnih točaka snimanja prostora senzorima na mobilnom robotu koji je računski prihvatljiv za istraživanje nepoznatih velikih unutarnjih prostora. - algoritam za izdvajanje manjih zatvorenih prostornih cjelina tijekom istraživanja prostora koji omogućuje postupno istraživanje nepoznatog velikog unutarnjeg prostora. - algoritam za planiranje putanja robota koje prolaze dijelovima prostora koji povećavaju točnost

Abstract (english)

This thesis introduces exploration of large indoor environments for dense 3D thermal model building by a mobile robot. The exploration strategy refers to obtaining exact, discrete scanning positions from which an unknown environment is completely explored. To overcome the problem of enormous memory consumption while exploring the environments and to reduce the computational effort the exploration strategy combines 2D and 3D exploration approach, taking advantages of the both approaches. Room detection algorithm divides the environment into enclosed spaces and thus enables switching between 2D and 3D exploration. In order to improve robot's localization while autonomously exploring an unknown environment the robot needs to revisit positions seen before. An active SLAM algorithm that continuously improves robot’s localization while moving towards a goal position was developed. The path planning is based on the D* shortest path graph search algorithm with negative edge weights for finding the shortest path taking into account localization uncertainty. The following scientific contribution is the result of the thesis: - exploration algorithm, for obtaining discrete scanning positions, which is computationally acceptable for large indoor environments - room detection algorithm while environment exploration for sequential exploration of large indoor environments

• Keywords: istraživanje nepoznatog prostora; mobilni robot; izdvajanje sobe; planiranje putanje; toplinski model prostora; laserski senzor udaljenosti; lokalizacija; SLAM
• Keywords (english): exploration of unknown environments; mobile robots; room detection; path planning; thermal model of environments; laser range sensor; localization; active SLAM
• Language: croatian
• URN:NBN: urn:nbn:hr:168:736982
• Study programme: Title: Electrical Engineering and Computing; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti elektrotehnike i računarstva (Doktor znanosti elektrotehnike i računarstva)
• Type of resource: Text
• Extent: 124 str. ; 30 cm.
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-04-05 13:34:25