The goal of this project is to develop a standard component that enables mobile devices such as PDAs, mobile phones, or navigation systems to collect, process, and use collected past routes for future route prediction.
Imagine that you are a car owner and consider the driving pattern for yourself during a normal week. The pattern is, given the time and weekday, probably the same each week with some minor deviations.
A short paper is available that presents an approach for online route prediction based on historical route data. The model used is, however, rather complex, and it would be beneficial to have a simpler and more practical version; perhaps at the expense of a less accurate system. The developed software will be integrated into the Streamspin mobile client and server kernel, but the component should be as general as possible, possibly to be released as an open source or freeware component.