Lokalizacija autonomnog vozila u simuliranom urbanom okruženju

master's thesis

• Title (english): Autonomous vehicle localization in a simulated urban environment
• Author: Matija Vukić
• Mentor: Ivan Marković (mentor)
• Committee member: Ivan Marković (predsjednik povjerenstva)
• Committee member: Ivan Petrović (član povjerenstva)
• Committee member: Marija Seder (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Postoje razni algoritmi za lokalizaciju vozila pomoću senzorskih očitanja. S ob-
zirom da ne postoji savršen algoritam rezultati odstupaju od stvarnih vrijednosti. Ko-
risteći simulator sa referentnim podacima možemo jednostavno usporediti stvarne po-
datke s rezultatima algoritama. Evaluiraju se generalizirajući algoritam iteracije naj-
bliže točke te takod̄er taj isti algoritam ali s prijašnjim filtriranjem podataka. Rezultati
su prikazani grafovima razlika koordinata i rotacija na poskupovima referentnih poda-
taka. Takod̄er su izračunate srednje pogreške i prikazane tablično zbog jednostavnije
usporedbe. Predstavljen je specifičan način evaluaciju algoritama korištenjem simu-
latora, ali čiji se koraci mogu općenito primjeniti na druge izvore podataka i druge
algoritme.

Abstract (english)

There are various vehicle location algorithms using sensory readings. Since there
is no perfect algorithm the results deviate from the actual values. Using a simulator
with reference data, we can easily compare actual data with algorithm results. Evalu-
ated algorithms are generalized iterative closest point and previous algorithm but with
the previous data filtering. The results are shown using graphs of the coordinate dif-
ferences and rotation differences between estimated values and referent values. Mean
errors were also calculated and displayed in the table. A specific way of evaluating
algorithms using a simulator was presented but whose steps generally be applied to
other data sources and other algorithms.

• Keywords: simulacija; lokalizacija; iteracija najbližih točaka; Carla; evaluacija; PCL; oblak točaka
• Keywords (english): simulation; localization; iterative closest point; Carla; evaluation; PCL; point cloud
• Language: croatian
• URN:NBN: urn:nbn:hr:168:063538
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-01-07 23:03:49