Factored extended Kalman filter for monocular vision-aided inertial navigation

Daniel Magree, Eric N. Johnson

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

This paper describes a novel visual simultaneous localization and mapping system based on the UD factored extended Kalman filter. A novel method for marginalizing and initializing state variables is presented, allowing for featurestobeadded and removed directlyto and from the covariance factors. An analysis of the number of operations in the marginalization and initialization algorithm are presented. A randomized analysis demonstrates the improvement in numerical stability over the standard and Joseph-form extended Kalman filter by as much as three orders of magnitude. The navigation system is implemented on an 80 kg unmanned aerial vehicle and a 0.5 kg unmanned aerial vehicle, and flight-test and simulation results with a controller in the loop are presented. The flight-test results agree with the simulation results and show a low state error and a consistent error covariance.

Original languageEnglish (US)
Pages (from-to)475-490
Number of pages16
JournalJournal of Aerospace Information Systems
Volume13
Issue number12
DOIs
StatePublished - Jan 1 2016

Fingerprint

Extended Kalman filters
Unmanned aerial vehicles (UAV)
Navigation
Convergence of numerical methods
Navigation systems
Controllers

All Science Journal Classification (ASJC) codes

  • Aerospace Engineering
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

@article{41868002ddda4d88a483ffbe2da4ee72,
title = "Factored extended Kalman filter for monocular vision-aided inertial navigation",
abstract = "This paper describes a novel visual simultaneous localization and mapping system based on the UD factored extended Kalman filter. A novel method for marginalizing and initializing state variables is presented, allowing for featurestobeadded and removed directlyto and from the covariance factors. An analysis of the number of operations in the marginalization and initialization algorithm are presented. A randomized analysis demonstrates the improvement in numerical stability over the standard and Joseph-form extended Kalman filter by as much as three orders of magnitude. The navigation system is implemented on an 80 kg unmanned aerial vehicle and a 0.5 kg unmanned aerial vehicle, and flight-test and simulation results with a controller in the loop are presented. The flight-test results agree with the simulation results and show a low state error and a consistent error covariance.",
author = "Daniel Magree and Johnson, {Eric N.}",
year = "2016",
month = "1",
day = "1",
doi = "10.2514/1.I010419",
language = "English (US)",
volume = "13",
pages = "475--490",
journal = "Journal of Aerospace Information Systems",
issn = "1542-9423",
publisher = "American Institute of Aeronautics and Astronautics Inc. (AIAA)",
number = "12",

}

Factored extended Kalman filter for monocular vision-aided inertial navigation. / Magree, Daniel; Johnson, Eric N.

In: Journal of Aerospace Information Systems, Vol. 13, No. 12, 01.01.2016, p. 475-490.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Factored extended Kalman filter for monocular vision-aided inertial navigation

AU - Magree, Daniel

AU - Johnson, Eric N.

PY - 2016/1/1

Y1 - 2016/1/1

N2 - This paper describes a novel visual simultaneous localization and mapping system based on the UD factored extended Kalman filter. A novel method for marginalizing and initializing state variables is presented, allowing for featurestobeadded and removed directlyto and from the covariance factors. An analysis of the number of operations in the marginalization and initialization algorithm are presented. A randomized analysis demonstrates the improvement in numerical stability over the standard and Joseph-form extended Kalman filter by as much as three orders of magnitude. The navigation system is implemented on an 80 kg unmanned aerial vehicle and a 0.5 kg unmanned aerial vehicle, and flight-test and simulation results with a controller in the loop are presented. The flight-test results agree with the simulation results and show a low state error and a consistent error covariance.

AB - This paper describes a novel visual simultaneous localization and mapping system based on the UD factored extended Kalman filter. A novel method for marginalizing and initializing state variables is presented, allowing for featurestobeadded and removed directlyto and from the covariance factors. An analysis of the number of operations in the marginalization and initialization algorithm are presented. A randomized analysis demonstrates the improvement in numerical stability over the standard and Joseph-form extended Kalman filter by as much as three orders of magnitude. The navigation system is implemented on an 80 kg unmanned aerial vehicle and a 0.5 kg unmanned aerial vehicle, and flight-test and simulation results with a controller in the loop are presented. The flight-test results agree with the simulation results and show a low state error and a consistent error covariance.

UR - http://www.scopus.com/inward/record.url?scp=85012159731&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85012159731&partnerID=8YFLogxK

U2 - 10.2514/1.I010419

DO - 10.2514/1.I010419

M3 - Article

AN - SCOPUS:85012159731

VL - 13

SP - 475

EP - 490

JO - Journal of Aerospace Information Systems

JF - Journal of Aerospace Information Systems

SN - 1542-9423

IS - 12

ER -