Acquisition and Use of Mobility Habits for Personal Assistants

With large parts of human population increasingly living in big cities, the mobility behavior of humans is about to change faster than ever before. Not only convenience and increasing ecological awareness lead to more intermodal mobility behavior, also the rise of new mobility options like car-or bike sharing are becoming more and more common. Wide distribution of smartphones and the on-trip availability of high-speed Internet let users inform themselves about a vast variety of mobility options. This information overload can overburden users who often have the simple wish to conveniently travel from A to B. Digital Mobility Assistants ease the burden of selecting the best mobility option for a particular user by incorporating the users' habits and preferences and providing relevant information at just the right time. To enable such intelligent assistance, we propose to create personalized mobility models that include not only information about habitual trips and destinations, but also allow for the detection of preferred travel modes. Our system is specifically designed to use sparse sensor data from mobile devices, such as smartphones, to offer an adequate balance between battery-life and data quality.