Multiple moving objects tracking based on random finite sets and Lie groups

doctoral thesis

• Title (croatian): Praćenje više gibajućih objekata zasnovano na slučajnim konačnim skupovima i Lievim grupama
• Author: Josip Ćesić; MBZ: -333478
• 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 - Electrical engineering

Abstract

Autonomous navigation of an agent strongly relies on the capability of tracking multiple moving objects. The first part of the thesis presents a method for multiple moving objects tracking on the unit sphere based on the von Mises distribution, and probability hypothesis density filter based on random finite sets. The state of objects in the agent's surrounding are typically determined with their position and orientation which evolve on a non-Euclidean geometry. The thesis further proposes several methods for estimating motion evolving on the special Euclidean group based on the extended Kalman filter on Lie groups. As an alternative to the extended Kalman filter on Lie groups, the thesis also proposes the estimation method relying on an information form for states evolving on matrix Lie groups. The thesis proposes two methods suitable for multiple objects tracking applications when states evolve on matrix Lie groups. The first one relies on joint integrated probabilistic data association filter, while the second one employs the probability hypothesis density filter on matrix Lie groups. Contributions: 1. Method for multiple moving objects tracking on the unit sphere based on the von Mises distribution and the random finite sets. 2. Method for moving object tracking in the space of the special Euclidean group based on the extended Kalman filter on Lie groups. 3. Extended information Kalman filter for state estimation on matrix Lie groups. 4. Method for multiple moving objects tracking on Lie Groups based on the concentrated Gaussian distribution and the random finite sets.

Abstract (croatian)

Autonomna navigacija agenta zasnovana je na sposobnostima praćenja više gibajućih objekata. Prvi dio rada bavi se problemom praćenja više gibajućih objekata na prostoru jedinične kružnice zasnovano na von Misesovoj distribuciji te filtru vjerojatnosti gustoće hipoteza zasnovanom na slučajnim konačnim skupovima. Stanja objekata u okolini mobilnoga agenta često su određena njegovom pozicijom i orijentacijom koja predstavlja primjer ne-Euklidske veličine. Disertacija nadalje predlaže nekoliko metoda za estimaciju gibanja na ne-Euklidskoj geometriji zasnovano na proširenom Kalmanovu filtru na Lievim grupama. Kao alternativa proširenom Kalmanovu filtru na Lievim grupama disertacija predlaže metodu estimacije zasnovanu na informacijskoj formi za stanja opisana Lievim grupama. Disertacija predlaže i dvije metode prikladne za praćenje više gibajućih objekata kada je stanje sustava opisano Lievim grupama. Prva metoda zasnovana je na filtru združenog integriranog vjerojatnosnog pridruživanja podataka, dok je drugi zasnovan na filtru vjerojatnosti gustoće hipoteza. Doprinosi: 1. Metoda praćenja više gibajućih objekata na jediničnoj sferi na temelju mjerenja smjera zasnovana na von Misesovoj razdiobi i slučajnim konačnim skupovima. 2. Metoda praćenja objekta u prostoru specijalne euklidske grupe zasnovana na proširenom Kalmanovu filtru na Lievim grupama. 3. Prošireni informacijski Kalmanov filtar za estimaciju stanja na matričnim Lievim grupama. 4. Metoda praćenja više gibajućih objekata na Lievim grupama zasnovana na koncentriranoj Gaussovoj razdiobi i slučajnim konačnim skupovima.

• Keywords: multiple moving objects tracking; Lie groups; directional statistics; concentrated Gaussian distribution; extended Kalman lter; extended information lter; random nite sets; probability hypothesis density; joint integrated probabilistic data association
• Keywords (croatian): pracenje više gibajucih objekata; Lieve grupe; usmjerena statistika; koncentrirana Gaussova razdioba; prošireni Kalmanov ltar; prošireni informacijski ltar; slucajni konacni skupovi; ltar gustoce vjerojatnosti hipoteza; združeno integrirano vjerojatnosno pridruživanje podataka
• Language: english
• URN:NBN: urn:nbn:hr:168:252322
• 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: 140 str. ; 30 cm.
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-04-04 13:42:43