Incremental Navigation Map Enhancement with GPS Tracks from Heterogeneous Sources
M. Vesely, C. Novak, A. Reh, H. Mayr - Incremental Navigation Map Enhancement with GPS Tracks from Heterogeneous Sources - Proceedings of The 2008 Internationa Conference on Machine Learning; Models, Technologies and Applications, Las Vegas, United States of America, 2008, pp. 787-793
This contribution describes our approach to incremental navigation map generation and enhancement based on Global Positioning System (GPS) track data gathered through tracing the paths of entities with heterogeneous mobility models (e.g., pedestrians, bicyclists, cars, etc.).
Our generic prototypical implementation is able to generate navigable maps from scratch as well as to enhance existing maps by deriving changes in the original map from the GPS tracks input, and to offer map updates for public download. For the core functionality of incremental semi-automatic map
learning, a proprietary graph matching algorithm is combined with a set of models that take the varieties of input data into consideration.
Areas of application comprise the continuous enhancement of navigation systems, their personalized adaptation, optimization of fleet management, and the generation of integral harmonized navigation data from heterogeneous sources.