Successfully reported this slideshow.
AGILE 2013 – Leuven, May 14-17, 2013Reconstructing movement tracesthrough a hybrid map-matchingalgorithmUnderstanding Urba...
AGILE 2013 – Leuven, May 14-17, 2013Research Framework• Branding slogan: “Plan your nicest route”– Specific information ab...
AGILE 2013 – Leuven, May 14-17, 2013Research Framework• PhD research:– “Mapping Linear Landscapes - Geosemanticmethods for...
AGILE 2013 – Leuven, May 14-17, 2013Algorithm
AGILE 2013 – Leuven, May 14-17, 2013Algorithm
AGILE 2013 – Leuven, May 14-17, 2013Algorithm
AGILE 2013 – Leuven, May 14-17, 2013Algorithm• Software: FME / Python/OSRM Routing Engine
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Hybrid map-matching algorithm• Geographical: Point Search Algorithm• Semanti...
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Vector Database:– TomTom• Preprocessing steps twofold:Routable dataset Point...
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Raw dataset– Create plausible trips per PERSONID• Time between registrations...
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Raw dataset– Analyse and filter trips on time passed andmeanspeed
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Raw dataset– Resulting trips (3302)• PersonID• TripID• Starttime• Endtime• M...
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Point Search Algorithm– Detect unambigious points along a trace:• Dual carri...
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Point Search Algorithm– Variable parameters in function of quality:• Dynamic...
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Similarity measure– A is routed segment– B is original segment of traceAB
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Similarity measures (Quality dependend threshold)– Frèchet distance– Relativ...
AGILE 2013 – Leuven, May 14-17, 2013Workflow• Hybrid map-matching algorithm
AGILE 2013 – Leuven, May 14-17, 2013Result: overviewGood similarityBad similarity
AGILE 2013 – Leuven, May 14-17, 2013Result: overviewGood similarity Bad similarity
AGILE 2013 – Leuven, May 14-17, 2013Result: comparison• Individual tripsCreated TripTRIPID: 3034Meanspeed: 4.88Person : 20...
AGILE 2013 – Leuven, May 14-17, 2013Result: comparison• Individual tripsCreated TripTRIPID: 1328Meanspeed: 6.9955Person : ...
AGILE 2013 – Leuven, May 14-17, 2013Result: comparison• Heatmap of ‘additional cycle trips’ (Raster)MaxMinRemark: Pixelsiz...
AGILE 2013 – Leuven, May 14-17, 2013Result: comparison• Vector dataset with continuous color on countMaxMin
AGILE 2013 – Leuven, May 14-17, 2013Result: comparison
AGILE 2013 – Leuven, May 14-17, 2013Discussion and Research Outlook• Customize routable data to specific activity– Exclude...
AGILE 2013 – Leuven, May 14-17, 2013Q&A
Upcoming SlideShare
Loading in …5
×

Reconstructing movement traces throug a hybrid map matching algorithm

820 views

Published on

Kevin Baker, Pascal Brackman, Philippe De Maeyer, Rik Van de Walle
University Ghent, Belgium; RouteYou, Belgium
Topic: “Reconstructing movement traces through a hybrid map-matching algorithm”

Published in: Technology, Education
  • Be the first to comment

Reconstructing movement traces throug a hybrid map matching algorithm

  1. 1. AGILE 2013 – Leuven, May 14-17, 2013Reconstructing movement tracesthrough a hybrid map-matchingalgorithmUnderstanding Urban Cycling: A DataChallengeKevin Baker
  2. 2. AGILE 2013 – Leuven, May 14-17, 2013Research Framework• Branding slogan: “Plan your nicest route”– Specific information about the road infrastructureand surroundings focused on his application
  3. 3. AGILE 2013 – Leuven, May 14-17, 2013Research Framework• PhD research:– “Mapping Linear Landscapes - Geosemanticmethods for information extraction, validationand enrichment using dynamic geodata”– intelligent aggregation and combination of novelgeographic information from a dynamiccommunity
  4. 4. AGILE 2013 – Leuven, May 14-17, 2013Algorithm
  5. 5. AGILE 2013 – Leuven, May 14-17, 2013Algorithm
  6. 6. AGILE 2013 – Leuven, May 14-17, 2013Algorithm
  7. 7. AGILE 2013 – Leuven, May 14-17, 2013Algorithm• Software: FME / Python/OSRM Routing Engine
  8. 8. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Hybrid map-matching algorithm• Geographical: Point Search Algorithm• Semantic: Attribute matching• Topological: Shortest Path Routing Engine
  9. 9. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Vector Database:– TomTom• Preprocessing steps twofold:Routable dataset Point cloud
  10. 10. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Raw dataset– Create plausible trips per PERSONID• Time between registrations < 5 minutes• Remove outliers/error:– HDOP < 5– Distance betweenregistrations (<2500 m)– outliers/error in Lat/Lon
  11. 11. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Raw dataset– Analyse and filter trips on time passed andmeanspeed
  12. 12. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Raw dataset– Resulting trips (3302)• PersonID• TripID• Starttime• Endtime• Meanspeed
  13. 13. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Point Search Algorithm– Detect unambigious points along a trace:• Dual carriage way• Parallel roads• Bearing difference• Analyse closestcandidates
  14. 14. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Point Search Algorithm– Variable parameters in function of quality:• Dynamic search distance (B)• Search interval (ΔA)• Allowed bearing difference• Number of closest candidates to analyse
  15. 15. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Similarity measure– A is routed segment– B is original segment of traceAB
  16. 16. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Similarity measures (Quality dependend threshold)– Frèchet distance– Relative/Absolute length difference– Area between segments– Turning function
  17. 17. AGILE 2013 – Leuven, May 14-17, 2013Workflow• Hybrid map-matching algorithm
  18. 18. AGILE 2013 – Leuven, May 14-17, 2013Result: overviewGood similarityBad similarity
  19. 19. AGILE 2013 – Leuven, May 14-17, 2013Result: overviewGood similarity Bad similarity
  20. 20. AGILE 2013 – Leuven, May 14-17, 2013Result: comparison• Individual tripsCreated TripTRIPID: 3034Meanspeed: 4.88Person : 207Traveltime: 0:36:36Additional trailTRIPID: 207-961Person : 207Traveltime: 0:16:00
  21. 21. AGILE 2013 – Leuven, May 14-17, 2013Result: comparison• Individual tripsCreated TripTRIPID: 1328Meanspeed: 6.9955Person : 146Traveltime: 0:49:25Additional trailTRIPID: 146-964Person : 146Traveltime: 0:15:00
  22. 22. AGILE 2013 – Leuven, May 14-17, 2013Result: comparison• Heatmap of ‘additional cycle trips’ (Raster)MaxMinRemark: Pixelsize of 4m
  23. 23. AGILE 2013 – Leuven, May 14-17, 2013Result: comparison• Vector dataset with continuous color on countMaxMin
  24. 24. AGILE 2013 – Leuven, May 14-17, 2013Result: comparison
  25. 25. AGILE 2013 – Leuven, May 14-17, 2013Discussion and Research Outlook• Customize routable data to specific activity– Exclude specific highwayclasses?– Resistance on edges (fix vs variable)?• Extensive post processing of delta (red lines)– How far can the automatic integration go?– Crowdsourcing/Outsourcing?• Fine-tuning the Point Search Algorithm• Extend similarity measure and thresholds• Resulting trip parameters
  26. 26. AGILE 2013 – Leuven, May 14-17, 2013Q&A

×