Odometrija mobilnoga robota zasnovana na radarskom senzoru

undergraduate thesis

• Title (english): Mobile robot odometry based on a radar sensor
• Author: Vlaho-Josip Štironja
• Mentor: Ivan Marković (mentor)
• Committee member: Ivan Marković (predsjednik povjerenstva)
• Committee member: Juraj Peršić (član povjerenstva)
• Committee member: Filip Marić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-08, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Odometrija je estimiranje vlastite pozicije u prostoru i vremenu na temelju podataka dobivenih od senzora. Često je izvedena kroz fuziju radara i kamere kod automobila. Prije određivanje vlastite pozicije potrebno je poznavati poziciju objekata u okolini, što se naziva praćenje meta. Za pridruživanje detekcija track-u potrebno je razaznavati stvarnu detekciju od šuma. U tu svrhu se koristi metoda klasteriranja koja generira klastere. Klasteri su grupa pojedinih podataka koja predstavljaju jedan objekt. Za završni rad se za dobivanje odometrije koristi tracking algoritam GNN dok za klasteriranje DBSCAN. Potrebno je implementirati algoritme i omogućiti komunikaciju putem TI AWR1843BOOST radara putem ROS-a te usporediti rezultate sa ground truth odometrijom.

Abstract (english)

Odometry is a term for estimating own position in space and time based on data from sensors. It is often performed through radar-camera fusion. Before calculating the position it is necessary to recognize the position of objects in environment (target tracking). Data association requires distinction between real detection and clutter. For this purpose clustering methods are used . Cluster is group of data that represents one object. This paper presents an odometry based on GNN algorithm (tracking algorithm) and DBSCAN (clustering algorithm). It is required to implement listed algorithms and enable communication via ROS to TI AWR1843BOOST radar and compare results with ground truth odometry.

• Keywords: odometrija; radar; klasteriranje; praćenje meta; DBSCAN; GNN
• Keywords (english): odometry; tracking; clustering algorithms; radar; DBSCAN; GNN
• Language: croatian
• URN:NBN: urn:nbn:hr:168:122318
• 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: 2023-01-13 16:19:13