kth.sePublications
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
Road Three-phase Traffic Modeling And Classification: using Traffic Cellular Automaton and Graph Neural Networks
KTH, School of Electrical Engineering and Computer Science (EECS).
2024 (English)Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
Abstract [en]

Traffic control and optimization have become increasingly critical with the rise of autonomous vehicles and the growing number of cars on the road. Effective traffic management requires a comprehensive understanding of road conditions, often achieved through traffic flow or density predictions using sensors or cameras. However, these data sources are frequently limited and noisy. This thesis proposes a novel approach using a traffic cellular automaton model to generate synthetic data representing various road conditions, classified into three phases according to the three-phase theory. We trained Graph Neural Networks on these synthetic, labeled graphs to predict the state of each road segment over time. The model demonstrated high performance, achieving 93% accuracy and excellent class/phase separation. This work contributes to traffic control by developing a strong and efficient traffic classification model and generating balanced, comprehensive data for improved traffic management.

Abstract [sv]

Trafikstyrning och optimering har blivit allt viktigare med framväxten av autonoma fordon och det ökande antalet bilar på vägarna. Effektiv trafikhantering kräver en omfattande förståelse för vägförhållandena, vilket ofta uppnås genom trafikflödeseller täthetsprognoser med hjälp av sensorer eller kameror. Dessa datakällor är dock ofta begränsade och brusiga. Denna avhandling föreslår en ny metod som använder en trafikcellulär automatonmodell för att generera syntetiska data som representerar olika vägförhållanden, klassificerade i tre faser enligt trefasteorin. Vi tränade grafneurala nätverk på dessa syntetiska, märkta grafer för att förutsäga tillståndet för varje vägsegment över tid. Modellen visade stark prestanda med 93% noggrannhet och utmärkt klass-/fas-separation. Detta arbete bidrar till trafikstyrning genom att utveckla en robust trafikklassificeringsmodell och generera balanserade, omfattande data för förbättrad trafikhantering.

Place, publisher, year, edition, pages
2024. , p. 55
Series
TRITA-EECS-EX ; 2024:839
Keywords [en]
Traffic Control, Three-phase Traffic Theory, Traffic Cellular Automaton, Synthetic Data, Graph Neural Networks
Keywords [sv]
Trafikstyrning, Three-phase Traffic Theory, Traffic Cellular Automaton, Syntetiska data, Graph Neural Networks
National Category
Electrical Engineering, Electronic Engineering, Information Engineering
Identifiers
URN: urn:nbn:se:kth:diva-360495OAI: oai:DiVA.org:kth-360495DiVA, id: diva2:1940459
Supervisors
Examiners
Available from: 2025-03-03 Created: 2025-02-26 Last updated: 2025-03-03Bibliographically approved

Open Access in DiVA

fulltext(7810 kB)28 downloads
File information
File name FULLTEXT01.pdfFile size 7810 kBChecksum SHA-512
a29e566b390691695acf675138541dad48a69bd2358e02f0c37b3f29c0c188f9da68f84f4039e649ef28e0070df773641f9e6f1dbfe182e83887382ae81b68f5
Type fulltextMimetype application/pdf

By organisation
School of Electrical Engineering and Computer Science (EECS)
Electrical Engineering, Electronic Engineering, Information Engineering

Search outside of DiVA

GoogleGoogle Scholar
Total: 28 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: 390 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