Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
A study of EKF-SLAM for vehicle state estimation.
KTH, School of Industrial Engineering and Management (ITM).
2019 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesisAlternative title
Studie av EKF-SLAM för tillstånds estimering av fordon. (Swedish)
Abstract [en]

Self driving vehicles consists of advanced Commercial-Off-The-Shelf components which are developed separately by different manufacturers and can introduce uncertainties to the system. To guarantee that the system meets the safety requirements is a complex task and it is important to test the performance of components beforehand. This paper examines a state estimation technique called EKF-SLAM and the aim of the research is to test the algorithm against erroneous sources and discuss its robustness. It is tested in a simulated environment called preScan against noise from sensors, timing delays and dynamic objects. In conclusion the performance of EKF seem to be limited and is dependent on a highly accurate feature extraction method. It showed to be robust against large sensor noise but for small errors a simpler method called odometry seemed much more preferable. Since EKF is derived from the assumption that all landmarks are static it became a complex task to cope with dynamic objects, and with network delays the results show quite clearly that some compensations are needed.

Abstract [sv]

Självkörande fordon består av avancerade Commercial-Off-The-Shelf komponenter som utvecklas separat och introducerar osäkerheter i systemet. Att garantera att systemet ska uppfylla alla säkerhetskrav är en komplicerad uppgift och det är därför viktigt att granska komponenternas prestanda i förhand. Den här rapporten undersöker en SLAM algorithm som kallas EKF och syftet med undersökning är att testa algoritmen mot felaktiga källor och diskutera dess robusthet. Algortihmen testas i en trafiksimulator som kallas preScan och testas mot oljud i hastighetsmätaren, rörliga objekt och tidsfördröjningar i ett distribuerat nätverk. Sammanfattningsvis så finns det klara begränsningar i hur bra EKF kan prestera och är beroende av en bra sensor model. Algoritmen visade sig vara robust mot stora fel i hastighetsmätaren, men för små fel tycks en enkel distansmätnings-metod vara mycket mer föredragen. Eftersom EKF härleds från antagandet att alla objekt är statiska så blev det en komplex uppgift att hantera dynamiska objekt. Tidsfördröjningar i nätverket påverkar resultaten tydligt och visar på att kompensationer behöver göras.

Place, publisher, year, edition, pages
2019. , p. 67
Series
TRITA-ITM-EX ; 2019:400
National Category
Engineering and Technology
Identifiers
URN: urn:nbn:se:kth:diva-263289OAI: oai:DiVA.org:kth-263289DiVA, id: diva2:1367927
Supervisors
Examiners
Available from: 2019-11-05 Created: 2019-11-05 Last updated: 2019-11-05Bibliographically approved

Open Access in DiVA

fulltext(7843 kB)23 downloads
File information
File name FULLTEXT01.pdfFile size 7843 kBChecksum SHA-512
becfd5362b1464617d5deae7adc1f8d04aef85bbc2aef6afdb7503dab9e527172a97441034340819c67e1c49c1f0a4a764261cc29d28a433fdce8cc9db9c40c5
Type fulltextMimetype application/pdf

By organisation
School of Industrial Engineering and Management (ITM)
Engineering and Technology

Search outside of DiVA

GoogleGoogle Scholar
Total: 23 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 28 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf