Realizacija SLAM sustava uz algoritam otvorenog koda i LiDAR čvrstog stanja

master's thesis

• Title (english): Implementation of SLAM System with the Open_Source Algorithm and Solid -State LiDAR
• Author: Nino Franković
• Mentor: Loris Redovniković (mentor)
• Committee member: Loris Redovniković (predsjednik povjerenstva)
• Committee member: Đuro Barković (član povjerenstva)
• Committee member: Mladen Zrinjski (član povjerenstva)
• Granter: University of Zagreb; Faculty of Geodesy; Zagreb
• Defense date and country: 2022-02-11, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Geodesy

Abstract

LiDAR je mjerni uređaj koji pomoću laserskih zraka opaža objekte i pojave u svojoj
okolini. Na ovaj način dobiva se veliki obujam informacija o prostoru u znatno kraće vrijeme
no što je to bilo potrebno kod prethodnih metoda izmjere. Međutim, kako bi podaci dobiveni
laserskim skenerima bili korisni, potrebno je dobro poznavati problematiku samog načina
nastanka, obrade i pohrane trodimenzionalnog oblaka točaka. Ako se tome doda i problem
kretanja, te potreba za praćenjem lokacije uređaja za svo vrijeme snimanja, dolazi se do
kompleksnog sustava koji uključuje razne mjerne sustave i softverska rješenja. Kartiranje
okoliša pri kojem se svo vrijeme vrši i lokalizacija naziva se SLAM i zbog svoje kompleksnosti
cijena gotovih komercijalnih SLAM sustava je vrlo visoka. Cilj ovog rada je upogoniti SLAM
sustav na način da se koristi niskobudžetni LiDAR čvrstog stanja, a kao softverska podrška,
otvoreni algoritam koji je svima dostupan na GitHub platformi. Korišteni LiDAR uređaj je
Livox Mid-100, a korišteni algoritam kao softverska podrška je razvijen na sveučilištu u Hong
Kongu pod nazivom LOAM Livox. Eventualna mogućnost budućeg korištenja ovakvih sustava
u geodetske ili slične svrhe snimanja okoliša zasigurno bi približila lasersku tehnologiju širem
broju korisnika. Korištenje cjenovno prihvatljivijih novih tehnologija također bi pozitivno
utjecalo na modernizaciju struke općenito.

Abstract (english)

LiDAR is a measuring device that uses laser beams to record objects and
phenomena in its environment. In this way, a large amount of spatial information is
obtained in a considerably shorter time than was required with previous measurement
methods. However, in order for the data obtained by laser scanners to be useful, it is
necessary to have an understanding of the issues of the very way of creating, storing and
processing the so-called 3D point clouds. Adding the problem of movement, as well as
the need to monitor the location of the device for the entire recording time, results is a
complex system that includes various measuring devices and software solutions. Mapping
the environment while simultaneously preforming localizations is named SLAM, and due
to its complexity the price of ready-made commercial SLAM systems is very high. The
aim of this study is to power up the SLAM system in a way that uses a low-budget solidstate LiDAR, and as software support, an open type algorithm that is publicly available
via the GitHub platform. The LiDAR device used is the Livox Mid-100, while the
algorithm used is LOAM Livox, developed at the University of Hong Kong. The possible
future use of such systems for geodetic or similar purposes of environmental mapping
would certainly bring laser technology closer to a wider number of users. The use of more
affordable new technologies would also have a positive impact on the modernization of
the profession in general.

• Keywords: LiDAR čvrstog stanja; SLAM; ROS; LOAM
• Keywords (english): Solid-state LiDAR; SLAM; ROS; LOAM
• Language: croatian
• URN:NBN: urn:nbn:hr:256:764157
• Study programme: Title: Geodesy and Geoinformatics; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka geodezije i geoinformatike (magistar/magistra inženjer/inženjerka geodezije i geoinformatike)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2022-02-16 09:20:53