Needs and potential of 3D city information and sensor fusion technologies for vehicle positioning in urban environments

Abstract

We proposed a concept for an absolute vehicle positioning architecture in urban environments by accomplishing 3D city data and car side laser scanner. Today there is no reliable technology for a long time stable and absolute positioning in urban environments. While GNSS suffers from outtakes, SLAM and dead reckoning technologies are only precise during a limited time span. Because car mounted laser scanner produce bevel cuts, real 3D façade information is needed to apply a feature extraction algorithm. Our approach is based on a web feature service connecting the vehicle to a 3D city database. Starting from an initial position, a spatial request is phased to the database and geometries are sent to the car. There, feature extraction and map matching algorithm support an absolute positioning.