DriveSense: Contextual handling of large-scale route map data for the automobile

Frederik Wiehr, Vidya Setlur, Alark Joshi

In: IEEE Conference Publications. IEEE International Conference on Big Data (IEEE BigData-2013) October 6-9 Santa Clara CA United States Seiten 87-94 IEEE 2013.


Automakers are increasingly providing connectivity enhancements for vehicles to download navigational data, as well as to upload sensor information to the cloud. Generally, while more data may be better, for the driver on-the-go, information needs to be displayed in a manner that can be comprehended rather quickly. One of the major problems with visualizing route maps is that the amount of information visualized is always the same regardless of the fact that an individual may be more familiar with the region or whether an individual is driving at varying speeds. Research has shown that complex visualizations with visual clutter can cause cognitive overload that adversely affects the performance of a user. Additionally, the attention and interaction abilities of a driver are significantly compromised in a vehicular environment. We propose DriveSense, a context-sensitive visualization system that automatically varies the GPS updates and the corresponding visualization being displayed to the user based on the speed of the vehicle as well as the familiarity of the region that the user is driving in. Based on a user evaluation, we found that subjects preferred using the automatic visualizations of route maps generated by DriveSense than the visual representations shown by a standard GPS. We also computed visual clutter for our visualizations at varying speeds and found that the clutter was significantly less for the routes displayed by DriveSense for faster speeds as compared to slower speeds.

Weitere Links

drivesense.pdf (pdf, 4 MB )

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence