EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Track-to-Track Fusion for Cooperative Perception Using Collective Perception Messages

Redge Melroy Castelino iD, Shrijal Pradhan iD, Axel Hahn iD

Sensors · 23.03.2026 · 26 (6) · 2003

DOI: 10.3390/s26062003 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 23.03.2026. Verfasst von Redge Melroy Castelino und 2 weiteren Autor:innen. Beteiligte Einrichtung: Institute of Systems Engineering for Future Mobility, German Aerospace Center (DLR), 26121 Oldenburg, Germany. Die Arbeit nennt 53 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Vehicle-to-everything communication grants connected and automated road vehicles the opportunity to share their sensor information such as detected road objects for collective awareness. This paper compares various state fusion strategies within a high-level cooperative perception architecture, focusing on the fusion of object-level information provided in standard Collective Perception Messages. This work compares five track-to-track fusion methods, namely Covariance Intersection, Inverse Covariance Intersection, Adapted Extended Kalman Filter, Adapted Unscented Kalman Filter and Information Matrix Fusion, using a simulation framework built with CARLA and Autoware. The methods are analyzed in a case study to assess their performance under different vehicle maneuvers and varying input information accuracy. The case study highlights trade-offs between fusion strategies and illustrate their behavior in asynchronous multi-agent scenarios. While the analysis is conducted in simulation, the architecture is designed to be extensible, and directions for future development are outlined, including the integration of classification and object confidence fusion modules.

Schlagwörter

Stichwörter der Autor:innen

collective perception messagescooperative perceptionhigh-level data fusiontrack-to-track fusionvehicle-to-everything

Autor:innen und Institutionen

  • Institute of Systems Engineering for Future Mobility, German Aerospace Center (DLR), 26121 Oldenburg, Germany.
  • Department of Computing Science, Carl von Ossietzky University, 26129 Oldenburg, Germany.

ORCID-Profile: Redge Melroy Castelino, Shrijal Pradhan, Axel Hahn

Zitieren

APA

Castelino, R. M., Pradhan, S., & Hahn, A. (2026). Track-to-Track Fusion for Cooperative Perception Using Collective Perception Messages. Sensors, 26(6), 2003. https://doi.org/10.3390/s26062003

Vancouver

Castelino RM, Pradhan S, Hahn A. Track-to-Track Fusion for Cooperative Perception Using Collective Perception Messages. Sensors. 2026;26(6):2003. doi:10.3390/s26062003

BibTeX
@article{Castelino2026,
  author = {Castelino, Redge Melroy and Pradhan, Shrijal and Hahn, Axel},
  title = {Track-to-Track Fusion for Cooperative Perception Using Collective Perception Messages},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {6},
  pages = {2003},
  doi = {10.3390/s26062003}
}
Kostenlos starten →