The paper deals with the fusion of multiobject information over a network of heterogeneous and geographically dispersed nodes with sensing, communication and processing capabilities. To exploit the benefits of sensor networks for multiobject estimation problems, like e.g. multitarget tracking and multirobot SLAM (Simultaneous Localization and Mapping), a key issue to be addressed is how to consistently fuse (average) locally updated multiobject densities. In this paper we discuss the generalization of Kullback-Leibler average, originally conceived for single-object densities (i.e. probability density functions) to (both unlabeled and labeled) multiobject densities. Then, with a view to develop scalable and reliable distributed multiobject esti- mation algorithms, we review approaches to iteratively compute, in each node of the network, the collective multiobject average via scalable and neighborwise computations.

Average Kullback-Leibler divergence for random finite sets / Battistelli, Giorgio; Chisci, Luigi; Fantacci, Claudio; Farina, Alfonso; Vo, Ba-Ngu. - STAMPA. - (2015), pp. 1359-1366. (Intervento presentato al convegno 18th International Conference on Information Fusion tenutosi a Washington, DC, USA nel Just 6-9, 2015).

Average Kullback-Leibler divergence for random finite sets

BATTISTELLI, GIORGIO;CHISCI, LUIGI;FANTACCI, CLAUDIO;
2015

Abstract

The paper deals with the fusion of multiobject information over a network of heterogeneous and geographically dispersed nodes with sensing, communication and processing capabilities. To exploit the benefits of sensor networks for multiobject estimation problems, like e.g. multitarget tracking and multirobot SLAM (Simultaneous Localization and Mapping), a key issue to be addressed is how to consistently fuse (average) locally updated multiobject densities. In this paper we discuss the generalization of Kullback-Leibler average, originally conceived for single-object densities (i.e. probability density functions) to (both unlabeled and labeled) multiobject densities. Then, with a view to develop scalable and reliable distributed multiobject esti- mation algorithms, we review approaches to iteratively compute, in each node of the network, the collective multiobject average via scalable and neighborwise computations.
2015
Proceedings 18th International Conference on Information Fusion
18th International Conference on Information Fusion
Washington, DC, USA
Just 6-9, 2015
Battistelli, Giorgio; Chisci, Luigi; Fantacci, Claudio; Farina, Alfonso; Vo, Ba-Ngu
File in questo prodotto:
File Dimensione Formato  
FUSION2015.pdf

Accesso chiuso

Descrizione: Articolo principale
Tipologia: Pdf editoriale (Version of record)
Licenza: Tutti i diritti riservati
Dimensione 320.56 kB
Formato Adobe PDF
320.56 kB Adobe PDF   Richiedi una copia

I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificatore per citare o creare un link a questa risorsa: https://hdl.handle.net/2158/1075191
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 23
  • ???jsp.display-item.citation.isi??? 16
social impact