Title: Multiple target tracking using particle filtering and multiple model for manoeuvring targets
Authors: Zahir Messaoudi; Mourad Oussalah
Addresses: School of Electronics, Electrical and System Engineering, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK; Military Polytechnique School of Algiers, Bordj El Bahri, Algeria ' School of Electronics, Electrical and System Engineering, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
Abstract: This paper addresses the problem of data fusion for joint multiple manoeuvring target tracking and classification where centralised versus decentralised architectures are contrasted. The proposal advocates the use of a hybrid approach combining a particle filter (PF) like method to deal with system nonlinearities, Fitzgerald's cheap joint probabilistic data association filter (CJPDAF) for the purpose of data association/target estimation problems and multiple model filter (MMF) to handle the issue of target manoeuvrability, yielding CJPDA-MMPF algorithm. The feasibility and the performances of the proposal have been demonstrated using complex scenarios involving a set of Monte Carlo simulations dealing with four crossing targets with different operations modes. Constant velocity model and constant turn rate model were employed to generate the manoeuvres. The results of the simulations demonstrate the feasibility and the superiority of the centralised fusion architecture approach when combined with multiple model particle filter and cheap joint data association filter.
Keywords: data fusion; multiple targets; manoeuvring targets; target tracking; particle filter; cheap JPDA filter; joint probabilistic data association; MMPF; modelling; centralised architectures; decentralised architectures; target estimation; multiple model filter; target manoeuvrability; Monte Carlo simulation; constant velocity; constant turn rate.
DOI: 10.1504/IJAAC.2015.073482
International Journal of Automation and Control, 2015 Vol.9 No.4, pp.303 - 332
Received: 14 Sep 2015
Accepted: 15 Sep 2015
Published online: 09 Dec 2015 *