SMAP: A Map Extension Framework for Intelligent Travel Systems

In the last few decades, special attention has been drawn to the autonomous vehicles and smart travel system in order to provide safer roads and a more comfortable driving experience. Acquiring knowledge about the locations type (home, work, shop, restaurant, etc.) and the possible pause time helps in transforming the vehicles into smart entities having similar road view as the human drivers. This understanding results subsequently in deriving more accurate destination predictions without the driver intervention and providing consequently more sophisticated high level applications. This paper proposes an Openstreetmap based framework (SMAP) for mining the significant locations serving as possible points of interest (POI). Moreover, it develops a semantic annotation technique to automatically annotate the POIs with category tags serving as a prerequisite for locations search and recommendation services. Furthermore, the work addresses the parking problem by assigning each POI a set of the nearest parking locations.