Tracking of Cooperative and Non-Cooperative UAVs with Labeled Multi-Bernoulli Filter

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

This paper addresses the problem of tracking cooperative and non-cooperative Unmanned Aerial Vehicles (UAVs) in low-altitude environments using a novel Generalized Labeled Multi-Bernoulli (GLMB) filter. First, we propose an enhanced birth model that incorporates real-time position information transmitted by cooperative UAVs. Second, we introduce distinct label management strategies for cooperative and non-cooperative targets, significantly improving trajectory estimation accuracy for non-cooperative UAVs. Third, we develop a distributed sensor fusion framework that integrates local sensor measurements with cooperative UAV data to enhance tracking performance for both target types. Simulation experiments demonstrate the proposed filter's superior tracking capabilities in mixed UAV operational scenarios, particularly in maintaining continuous tracks.

Original languageEnglish
Title of host publicationProceedings of 2025 IEEE 15th International Conference on Signal Processing, Communications and Computing, ICSPCC 2025
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9798331565466
DOIs
StatePublished - 2025
Event15th IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2025 - Hong Kong, China
Duration: 18 Jul 202521 Jul 2025

Publication series

NameProceedings of 2025 IEEE 15th International Conference on Signal Processing, Communications and Computing, ICSPCC 2025

Conference

Conference15th IEEE International Conference on Signal Processing, Communications and Computing, ICSPCC 2025
Country/TerritoryChina
CityHong Kong
Period18/07/2521/07/25

Keywords

  • Arithmetic Average Fusion
  • Labeled Multi-Bernoulli Filter
  • Multi-Sensor Fusion
  • UAV Tracking

Fingerprint

Dive into the research topics of 'Tracking of Cooperative and Non-Cooperative UAVs with Labeled Multi-Bernoulli Filter'. Together they form a unique fingerprint.

Cite this