Generating Navigation Capable Maps from User Provided Data with Woodtracker

Common car navigation systems only provide routes within the road network. Dirt roads are out of the scope of these widespread systems. In this paper we present Woodtracker, an approach to provide off-road navigation for the timber industry of the federal state of Brandenburg. The basic principle of Woodtracker is that users of the system are producers as well as consumers of navigation information: Trucks driving through the woodlands record their tracks using a GPS based mobile device. This tracking data is transmitted to the woodtracker server. Within the Woodtracker server navigation data is represented as a graph with nodes and edges. New tracking data is incorporated in the graph, by identifying new roads (edges) as well as crossroads (nodes). The mobile client may also send route guidance requests to the server, which sends route information based on the stored navigation data. With an increasing number of users the system will deliver better navigation data. Further the system is able to adapt automatically to environment changes, e. g. new or closed roads. The mobile client of the prototype is implemented on the Android platform, whereas the server system is build with Geodjango, a specifically adapted geo science version of the Python based web framework Django.