Home // GEOProcessing 2019, The Eleventh International Conference on Advanced Geographic Information Systems, Applications, and Services // View article

Automated Construction of Road Networks from GPS Tracks

Authors:
Weiping Yang

Keywords: vehicle GPS data tracking; automated road extraction; road network analysis; geospatial data mining and knowledge discovery; machine learning

Abstract:
This paper describes a framework for automating road networks using GPS (Global Positioning Systems) track measurements. Through observation and experiments on the data, it is decided that automating road networks is done in a two-step process. The first step is to identify intersections of roads, following an intersection model that also identifies and holds tracking groups leading to the first legs of incident roads. The second step, road segments incident to each intersection node will be iteratively discovered by moving probe lines perpendicular to the heading directions of the generated roads. Initial intersections are assessed through analysing turns of vehicle trajectories and characteristics pertinent to where roads meet. Statistical techniques are used on tracks in relation to probe lines to exclude outliers and to locate median positions as vertices of roads. The method described in this paper exploits topological and geometric measures about neighbourhood of roads and applies machine learning techniques that iteratively compute optimized results for these measures.

Pages: 35 to 41

Copyright: Copyright (c) IARIA, 2019

Publication date: February 24, 2019

Published in: conference

ISSN: 2308-393X

ISBN: 978-1-61208-687-3

Location: Athens, Greece

Dates: from February 24, 2019 to February 28, 2019