Izbjegavanje prepreka korištenjem algoritma potencijalnih polja uz kartu dobivenu iz Google Cartographer SLAM-a

undergraduate thesis

• Title (english): Potential field based obstacle avoidance by using Google Cartographer
• Author: Juraj Obradović
• Mentor: Stjepan Bogdan (mentor)
• Committee member: Stjepan Bogdan (predsjednik povjerenstva)
• Committee member: Juraj Oršulić (član povjerenstva)
• Committee member: Ana Batinović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

SLAM (Simultaneous localization and mapping) je računalni problem izgradnje mape nepoznatog područja te paralelno lokaliziranje robota. SLAM tehnologije koriste podatke iz senzora te grade sliku prostora oko sebe te se pozicioniraju u istom. Riješio je problem lokaliziranja i mapiranja u zatvorenim prostorijama pošto GPS tamo ne funkcionira. Ovaj rad se temelji na Google Cartographeru koji u svom rješenju nudi kvalitetan SLAM u 2D i 3D korištenjem više senzorskih konfiguracija.
5. listopada 2016, Google je objavio kod za svoj 2d i 3d SLAM. Algoritam daje impresivne rezultate s obzirom da je to real-time SLAM te ga je vrlo lako implementirati jer Google software dolazi sa integracijom u ROS (Robot operating sistem).
Razvijanje savršenog SLAM algoritma dovelo bi do najvećeg napretka u razvitku autonomnih vozila, bespilotnih letjelica, svemirskih tehnologija, robota u kućanstvu te podvodnih vozila.
Rad također implementira algoritam potencijalnih polja s kojim se koristi da bi izbjegavao prepreke.

Abstract (english)

SLAM (Simultaneous localization and mapping) is computational problem of building a map in unknown environment and robot localization. SLAM solved the problem of building a map in closed environment as factory floor where GPS doesn't work. This paper implements Google Cartographer which is high quality SLAM implemented on many different sensor configurations.
On 5. October. 2016 Google announced Cartographer. Algorithm gives impressive results in real-time SLAM and is easy to implement because it is integrated in ROS (Robot operating system). Developing perfect SLAM algorithm would lead to the biggest improvement in self-driving cars, unmanned aerial vehicles, autonomous underwater vehicles, space technologies.
This paper also implements potential fields algorithm for obstacle avoidance.

• Keywords: SLAM; Potencijalna polja; Cartographer; Drone; Velodyne senzor
• Keywords (english): SLAM; potential fields; Cartographer; Drone; Velodyne sensor
• Language: croatian
• URN:NBN: urn:nbn:hr:168:565529
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-01-07 20:47:53