Line of sight analysis for urban mobile applications: a photogrammetric approach

Although inter visibility analysis between a GPS receiver and visible satellites is crucial for any real-time route navigation, traditional, mission-planning software does not incorporate it – they simply assume that the earth is flat. This paper seeks to correct the situation by outlining an advanced, semi-global matching technique which derives a Digital Surface Model (DSM), depicting the topography model of the earth surface for any study area. Our approach uses a GIS, combined with a freely available Global Navigation Satellite System (GNSS), to accurately predict pseudo satellite locations before building three-dimensional models of buildings from their footprints. We exploit high-resolution satellite data (Cartosat-1 and Geoeye-1) along with high-density elevation information, and we ultimately reduce the need for full-scale surveys and so minimize costs and time. Moreover, our combining of 3D models with GPS-derived, ephemeris data pertaining to densely built-up areas constitutes a valuable input for real time, mobile applications like car navigation systems, personal navigation, fleet tracking, asset tracking and general, location-based services.