An intelligent PHD filter implementation for maneuvering multi-target tracking

Author

Xu, Benlian ; Xu, Huigang

Author_Institution

Sch. of Electr. & Autom. Eng., Changshu Inst. of Technol., Changshu, China

fYear

2010

fDate

15-17 June 2010

Firstpage

2306

Lastpage

2310

Abstract

An intelligent probability hypothesis density (PHD) filter, under the guidance of ant clustering behavior, is proposed and applied to estimate the time-varying number of maneuvering targets and their individual states. Our algorithm consists of rough and fine ant clustering behavior. The former forms approximation of PHD and yields the number of targets; whereas the latter is designed with the help of traditional ant colony optimization algorithm to extract the state of each target. Numerical simulations demonstrate tracking multiple-target capability of our proposed algorithm, and the obtained results are compared with the Sequential Monte Carlo (SMC) method under a given benchmark problem.

Keywords

Monte Carlo methods; filtering theory; optimisation; pattern clustering; probability; target tracking; SMC method; ant colony optimization; fine ant clustering behavior; intelligent PHD filter; intelligent probability hypothesis density filter; maneuvering multitarget tracking; numerical simulation; rough ant clustering behavior; sequential Monte Carlo method; Algorithm design and analysis; Ant colony optimization; Approximation algorithms; Clustering algorithms; Filters; Monte Carlo methods; Numerical simulation; Sliding mode control; State estimation; Target tracking; ant colony optimization; clustering; multi-target filtering; probability hypothesis density;

fLanguage

English

Publisher

ieee

Conference_Titel

Industrial Electronics and Applications (ICIEA), 2010 the 5th IEEE Conference on

Conference_Location

Taichung

Print_ISBN

978-1-4244-5045-9

Electronic_ISBN

978-1-4244-5046-6

Type

conf

DOI

10.1109/ICIEA.2010.5515517

Filename

5515517