Computational cost analysis of extended Kalman filter in simultaneous localization and mapping (EKF-SLAM) problem for autonomous vehicle

Saiful Bahri Samsuri, Hairi Zamzuri, Mohd Azizi Abdul Rahman, Saiful Amri Mazlan, Abdul Hadi Abd Rahman 

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Extended Kalman filter (EKF) based solution is one of the most popular techniques for solving simultaneous localization and mapping (SLAM) problem. However, previous research showed the implementation of EKF for SLAM suffered with high computational costs, which affect the performance in real time application. This paper investigates the computational cost performance of an EKF-SLAM algorithm. The analysis was done by time measurement on sub-step motion update and measurement update on EKF by considering the total numbers of landmarks and numerous setting on range observation distance. The analytical results show that as the number of landmarks or range observation distances increased, the computational cost in measurement update step required more computation time compare to motion update step. Furthermore, improvements are needed to optimize the computational cost for the update step.

Original languageEnglish
Pages (from-to)7764-7768
Number of pages5
JournalARPN Journal of Engineering and Applied Sciences
Volume10
Issue number17
Publication statusPublished - 1 Jan 2015
Externally publishedYes

Fingerprint

Extended Kalman filters
Costs
Time measurement

Keywords

  • Autonomous vehicle
  • Computational cost
  • Extended kalman filter
  • Slam

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Computational cost analysis of extended Kalman filter in simultaneous localization and mapping (EKF-SLAM) problem for autonomous vehicle. / Samsuri, Saiful Bahri; Zamzuri, Hairi; Abdul Rahman, Mohd Azizi; Mazlan, Saiful Amri; Abd Rahman , Abdul Hadi.

In: ARPN Journal of Engineering and Applied Sciences, Vol. 10, No. 17, 01.01.2015, p. 7764-7768.

Research output: Contribution to journalArticle

@article{3e29c291e54f4bd9bbb8689a3b2880a0,
title = "Computational cost analysis of extended Kalman filter in simultaneous localization and mapping (EKF-SLAM) problem for autonomous vehicle",
abstract = "Extended Kalman filter (EKF) based solution is one of the most popular techniques for solving simultaneous localization and mapping (SLAM) problem. However, previous research showed the implementation of EKF for SLAM suffered with high computational costs, which affect the performance in real time application. This paper investigates the computational cost performance of an EKF-SLAM algorithm. The analysis was done by time measurement on sub-step motion update and measurement update on EKF by considering the total numbers of landmarks and numerous setting on range observation distance. The analytical results show that as the number of landmarks or range observation distances increased, the computational cost in measurement update step required more computation time compare to motion update step. Furthermore, improvements are needed to optimize the computational cost for the update step.",
keywords = "Autonomous vehicle, Computational cost, Extended kalman filter, Slam",
author = "Samsuri, {Saiful Bahri} and Hairi Zamzuri and {Abdul Rahman}, {Mohd Azizi} and Mazlan, {Saiful Amri} and {Abd Rahman }, {Abdul Hadi}",
year = "2015",
month = "1",
day = "1",
language = "English",
volume = "10",
pages = "7764--7768",
journal = "ARPN Journal of Engineering and Applied Sciences",
issn = "1819-6608",
publisher = "Asian Research Publishing Network (ARPN)",
number = "17",

}

TY - JOUR

T1 - Computational cost analysis of extended Kalman filter in simultaneous localization and mapping (EKF-SLAM) problem for autonomous vehicle

AU - Samsuri, Saiful Bahri

AU - Zamzuri, Hairi

AU - Abdul Rahman, Mohd Azizi

AU - Mazlan, Saiful Amri

AU - Abd Rahman , Abdul Hadi

PY - 2015/1/1

Y1 - 2015/1/1

N2 - Extended Kalman filter (EKF) based solution is one of the most popular techniques for solving simultaneous localization and mapping (SLAM) problem. However, previous research showed the implementation of EKF for SLAM suffered with high computational costs, which affect the performance in real time application. This paper investigates the computational cost performance of an EKF-SLAM algorithm. The analysis was done by time measurement on sub-step motion update and measurement update on EKF by considering the total numbers of landmarks and numerous setting on range observation distance. The analytical results show that as the number of landmarks or range observation distances increased, the computational cost in measurement update step required more computation time compare to motion update step. Furthermore, improvements are needed to optimize the computational cost for the update step.

AB - Extended Kalman filter (EKF) based solution is one of the most popular techniques for solving simultaneous localization and mapping (SLAM) problem. However, previous research showed the implementation of EKF for SLAM suffered with high computational costs, which affect the performance in real time application. This paper investigates the computational cost performance of an EKF-SLAM algorithm. The analysis was done by time measurement on sub-step motion update and measurement update on EKF by considering the total numbers of landmarks and numerous setting on range observation distance. The analytical results show that as the number of landmarks or range observation distances increased, the computational cost in measurement update step required more computation time compare to motion update step. Furthermore, improvements are needed to optimize the computational cost for the update step.

KW - Autonomous vehicle

KW - Computational cost

KW - Extended kalman filter

KW - Slam

UR - http://www.scopus.com/inward/record.url?scp=84943329207&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84943329207&partnerID=8YFLogxK

M3 - Article

VL - 10

SP - 7764

EP - 7768

JO - ARPN Journal of Engineering and Applied Sciences

JF - ARPN Journal of Engineering and Applied Sciences

SN - 1819-6608

IS - 17

ER -