Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • 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
Path Planning for Autonomous Heavy DutyVehicles in Unknown Environments using GraphSearch
KTH, School of Computer Science and Communication (CSC).
2013 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

The vision of autonomous vehicles is becoming a reality. Advanced driver assistance systems have the potential to improve transportation safety and efficiency, and offer the possibility of environmental and economical gains. However, navigation and control of autonomous vehicles is an intricate task. The vehicle needs to make decisions and respond to situations on its own.

Graph-search based route planning has the potential to reduce replanning times and provides the end-user with feasible and optimized plans. This thesis adopts the real-time capable, re-planning DLite algorithm as a powerful planning workhorse. The algorithm conducts graph-search on a state lattice graph, and uses motion primitives to construct a plan. The route planner inherently handles inexact motion, which lends itself well in practice as the vehicle cannot be expected to follow instructions to the letter.

The route planner is able to plan in large and obstacle-laden cavelike terrains, and according to a cost map (e.g. a map pointing out fuelefficient routes), with an average re-planning time of less than 0.1 seconds. The state lattice graph is by construction feasible, which renders the planned path of the route planner physically possible for the vehicle to follow in practice.

Abstract [sv]

Autonoma fordon är en framtidsvision som aktualiseras allt mer. Avancerade förarsystem har potential att förbättra transportsäkerheten, och medför möjlighet till ekonomisk och miljömässig förbättring. Dock är navigering och kontroll av autonoma fordon en intrikat uppgift, då fordonet måste fatta beslut och hantera situationer på egen hand.

Grafbaserad ruttplanering minskar omplaneringstiderna, samtidigt som den genererar genomförbara och optimerade planer. Ruttplaneraren i denna avhandling nyttjar en realtidskapabel och kraftfull omplaneringsalgoritm vid namn DLite. Algoritmen utför grafsökningar i en regelbunden grafstruktur, och använder sig av smarta rörelsesegment då den konstruerar en rutt till mål. Ruttplaneraren tar hänsyn till oprecisa fordonsrörelser, vilket lämpar sig väl i praktiken eftersom fordonet inte kan förväntas följa instruktioner till punkt och pricka.

Ruttplaneraren klarar att planera i stora, hinderfyllda, och grottliknande terränger, samt i enlighet med en kostnadsfunktion (t.ex. en som pekar ut bränslesnåla rutter), med en genomsnittlig omplaneringtid på mindre än 0.1 sekunder. Grafstrukturen omfattar endast görliga rutter, vilket medför att den planerade rutten är fysiskt möjlig för fordonet att följa i praktiken.

Place, publisher, year, edition, pages
2013.
National Category
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-142402OAI: oai:DiVA.org:kth-142402DiVA: diva2:700130
Educational program
Master of Science in Engineering - Computer Science and Technology
Supervisors
Examiners
Available from: 2014-03-11 Created: 2014-03-03 Last updated: 2014-03-11Bibliographically approved

Open Access in DiVA

fulltext(7384 kB)521 downloads
File information
File name FULLTEXT01.pdfFile size 7384 kBChecksum SHA-512
71e0d1cbbe2fc920dad11fe1cd753135014d69d46c1a245762678bd6cfe51f01a37df252375fafd9760f8fa3bdcc197fb24e078732071e2a67e3021085dfd7ec
Type fulltextMimetype application/pdf

By organisation
School of Computer Science and Communication (CSC)
Computer Science

Search outside of DiVA

GoogleGoogle Scholar
Total: 521 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: 1472 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • 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