A fitted B-spline method to derive a representative 3D axis from a set of multiple road traces

Abstract

Collecting data to make an accurate representation for roads is an expensive process. Nevertheless, there is a collaborative alternative for this endeavour. Many drivers, bicyclists and even pedestrians have consumer-grade GPS (low precision) on their smartphones or electronic devices. Those users could transfer their road or track itineraries to a large database in order to compute the accurate geometry of any route. For each road or track, the large database would have many traces from which to infer an accurate representative 3D axis. Several inference methods have been proposed but most of them to fit the 2D trace data set. We propose to create a set of ordered points from the 3D trace data set and then using the least-squares method, to fit a B-spline curve to those points. The resulting parameterized curve will be a good representative 3D axis of the traces. Our method considers the nodes to be evenly separated and allows the system to recommend the number limit of nodes necessary to reach the convergence.

Keywords:

• Road
• B-spline
• multiple traces
• collaborative maps
• mining data

Disclosure statement

No potential conflict of interest was reported by the authors.