The linear opinion pool (LinOP) provides a potential solution to the problem of information fusion. However, the LinOP cannot be directly applied to multi-object fusion since the resulting fused multi-object density, in general, no longer belongs to the same family of the local ones, thus it cannot be utilized as prior informa- tion for the next recursion in Bayesian multi-object filtering. In this letter, by showing that the LinOP is actually the one that leads to minimum information loss (MIL), we propose to find the fused multi-object density that has the same form as the local ones and, at the same time, leads to MIL. The performance of MIL fusion is then compared with the one of the well-known generalized covariance intersection (GCI) fusion via simulations.

Multiobject fusion with minimum information loss / Lin Gao, Giorgio Battistelli, Luigi Chisci. - In: IEEE SIGNAL PROCESSING LETTERS. - ISSN 1070-9908. - STAMPA. - 27:(2020), pp. 201-205. [10.1109/LSP.2019.2963817]

Multiobject fusion with minimum information loss

Lin Gao;Giorgio Battistelli;Luigi Chisci
2020

Abstract

The linear opinion pool (LinOP) provides a potential solution to the problem of information fusion. However, the LinOP cannot be directly applied to multi-object fusion since the resulting fused multi-object density, in general, no longer belongs to the same family of the local ones, thus it cannot be utilized as prior informa- tion for the next recursion in Bayesian multi-object filtering. In this letter, by showing that the LinOP is actually the one that leads to minimum information loss (MIL), we propose to find the fused multi-object density that has the same form as the local ones and, at the same time, leads to MIL. The performance of MIL fusion is then compared with the one of the well-known generalized covariance intersection (GCI) fusion via simulations.
2020
27
201
205
Lin Gao, Giorgio Battistelli, Luigi Chisci
File in questo prodotto:
File Dimensione Formato  
IEEE-SPL-2020.pdf

Accesso chiuso

Descrizione: Articolo principale
Tipologia: Pdf editoriale (Version of record)
Licenza: DRM non definito
Dimensione 540.53 kB
Formato Adobe PDF
540.53 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/2158/1184524
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 50
  • ???jsp.display-item.citation.isi??? 45
social impact