User context estimation for public travel assistance and intelligent service scheduling

Despite the many advancements in the growing field of public transportation software, most applications are still stuck to a trip planning service model, where the information is provided to the users mainly before the trip begins. However, a more interesting paradigm exists, focusing on travel assistance, that is providing information to the users throughout the whole trip, much like private transport navigation systems. We argue that the main obstacle to the development of such applications is the difficulty in automatically reconstructing users contextual information, such the precise means (bus, train) he is traveling at any given moment. In this paper we present a system, based on particle filtering, for the estimation of the user context using only publicly available data and minimum sampling of his GPS position. The information estimated could be exploited for the design and the development of new kinds of applications for both users (e.g., real time travel assistance, context-aware games) and travel agencies (e.g., route analysis and assessment, overcrowding estimation).