Comparisons of PHD Filter and CPHD Filter for Space Object Tracking

Abstract

The Probability Hypothesis Density (PHD) filter and the Cardinalized PHD (CPHD) filter are two computationally tractable approximate Bayesian multiobject filters within the Finite Set Statistics framework. The PHD filter estimates the intensity function; the CPHD filter estimates the intensity function and the conditional distribution of the number of objects. The two filters are compared in an example of tracking three space objects, where the CPHD filter is shown to estimate the number of objects as well as the intensity function more accurately.

Meeting Name

2013 AAS/AIAA Astrodynamics Specialist Conference (2013: Aug. 11-15, Hilton Head, SC)

Department(s)

Mechanical and Aerospace Engineering

Keywords and Phrases

Estimation; Target tracking; Approximate Bayesian; Cardinalized PHD filter (CPHD); Conditional distribution; CPHD filters; Finite set statistics; Intensity functions; Probability hypothesis density filter; Space objects; Astrophysics

International Standard Book Number (ISBN)

978-0877036050

International Standard Serial Number (ISSN)

0065-3438

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2014 Univelt Inc., All rights reserved.

Publication Date

01 Jan 2014

This document is currently not available here.

Share

 
COinS