Multi-Sensor Space Debris Tracking for Space Situational Awareness With Labeled Random Finite Sets

Baishen Wei, Brett D. Nener

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

As a result of the dependence worldwide on satellite technology, it is now necessary to use advanced multi-target tracking algorithms for space debris tracking systems to maintain custody of space objects around the earth. One principal challenge is the correct association of observations with objects. This paper presents a multi-sensor, space-debris tracking algorithm using delta-generalized labeled multi-Bernoulli (delta-GLMB) filtering. The algorithm provides a solution to the key challenges (e.g., detection uncertainty, data association uncertainty, and clutter) in multiple object tracking. An efficient implementation of the multi-sensor delta-GLMB filter is proposed. In order to avoid exhaustively computing all the terms, we propose to use the ranked assignment algorithm with an extended assignment matrix for multiple sensors to determine the most significant terms. A measurement-based birth model is used to identify the previously unknown space objects. Sensors can have the same or different observation volumes. The expectation-maximization (EM) algorithm is used to approximate densities across observation volumes. The performance is demonstrated using the simulation results.

Original languageEnglish
Pages (from-to)36991-37003
Number of pages13
JournalIEEE Access
Volume7
DOIs
Publication statusPublished - 2019

Cite this

@article{93ae2dfddbf84d50a23c38031e95e312,
title = "Multi-Sensor Space Debris Tracking for Space Situational Awareness With Labeled Random Finite Sets",
abstract = "As a result of the dependence worldwide on satellite technology, it is now necessary to use advanced multi-target tracking algorithms for space debris tracking systems to maintain custody of space objects around the earth. One principal challenge is the correct association of observations with objects. This paper presents a multi-sensor, space-debris tracking algorithm using delta-generalized labeled multi-Bernoulli (delta-GLMB) filtering. The algorithm provides a solution to the key challenges (e.g., detection uncertainty, data association uncertainty, and clutter) in multiple object tracking. An efficient implementation of the multi-sensor delta-GLMB filter is proposed. In order to avoid exhaustively computing all the terms, we propose to use the ranked assignment algorithm with an extended assignment matrix for multiple sensors to determine the most significant terms. A measurement-based birth model is used to identify the previously unknown space objects. Sensors can have the same or different observation volumes. The expectation-maximization (EM) algorithm is used to approximate densities across observation volumes. The performance is demonstrated using the simulation results.",
keywords = "Space debris, delta-GLMB, multi-sensor, space situational awareness, MULTIAGENT SYSTEMS, VISUAL TRACKING, PROBABILITY, FILTER, CONSENSUS, ALGORITHM, ORDER, SLAM, CPHD",
author = "Baishen Wei and Nener, {Brett D.}",
year = "2019",
doi = "10.1109/ACCESS.2019.2904545",
language = "English",
volume = "7",
pages = "36991--37003",
journal = "IEEE Access",
issn = "2169-3536",
publisher = "IEEE, Institute of Electrical and Electronics Engineers",

}

Multi-Sensor Space Debris Tracking for Space Situational Awareness With Labeled Random Finite Sets. / Wei, Baishen; Nener, Brett D.

In: IEEE Access, Vol. 7, 2019, p. 36991-37003.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Multi-Sensor Space Debris Tracking for Space Situational Awareness With Labeled Random Finite Sets

AU - Wei, Baishen

AU - Nener, Brett D.

PY - 2019

Y1 - 2019

N2 - As a result of the dependence worldwide on satellite technology, it is now necessary to use advanced multi-target tracking algorithms for space debris tracking systems to maintain custody of space objects around the earth. One principal challenge is the correct association of observations with objects. This paper presents a multi-sensor, space-debris tracking algorithm using delta-generalized labeled multi-Bernoulli (delta-GLMB) filtering. The algorithm provides a solution to the key challenges (e.g., detection uncertainty, data association uncertainty, and clutter) in multiple object tracking. An efficient implementation of the multi-sensor delta-GLMB filter is proposed. In order to avoid exhaustively computing all the terms, we propose to use the ranked assignment algorithm with an extended assignment matrix for multiple sensors to determine the most significant terms. A measurement-based birth model is used to identify the previously unknown space objects. Sensors can have the same or different observation volumes. The expectation-maximization (EM) algorithm is used to approximate densities across observation volumes. The performance is demonstrated using the simulation results.

AB - As a result of the dependence worldwide on satellite technology, it is now necessary to use advanced multi-target tracking algorithms for space debris tracking systems to maintain custody of space objects around the earth. One principal challenge is the correct association of observations with objects. This paper presents a multi-sensor, space-debris tracking algorithm using delta-generalized labeled multi-Bernoulli (delta-GLMB) filtering. The algorithm provides a solution to the key challenges (e.g., detection uncertainty, data association uncertainty, and clutter) in multiple object tracking. An efficient implementation of the multi-sensor delta-GLMB filter is proposed. In order to avoid exhaustively computing all the terms, we propose to use the ranked assignment algorithm with an extended assignment matrix for multiple sensors to determine the most significant terms. A measurement-based birth model is used to identify the previously unknown space objects. Sensors can have the same or different observation volumes. The expectation-maximization (EM) algorithm is used to approximate densities across observation volumes. The performance is demonstrated using the simulation results.

KW - Space debris

KW - delta-GLMB

KW - multi-sensor

KW - space situational awareness

KW - MULTIAGENT SYSTEMS

KW - VISUAL TRACKING

KW - PROBABILITY

KW - FILTER

KW - CONSENSUS

KW - ALGORITHM

KW - ORDER

KW - SLAM

KW - CPHD

U2 - 10.1109/ACCESS.2019.2904545

DO - 10.1109/ACCESS.2019.2904545

M3 - Article

VL - 7

SP - 36991

EP - 37003

JO - IEEE Access

JF - IEEE Access

SN - 2169-3536

ER -