Assisting the Driver with Distance Estimation: Usability Evaluation of Graphical Presentation Alternatives for Local Traffic Events

Angela Mahr; Sandro Castronovo; Rafael Math; Christian Müller

In: Paula Kotzé; Gary Marsden; Gitte Lindgaard; Janet Wesson; Marco Winckler (Hrsg.). Human-Computer Interaction – INTERACT 2013. Pages 520-537, Lecture Notes in Computer Science (LNCS), Vol. 8119, ISBN 978-3-642-40476-4, Springer, Berlin - Heidelberg, 2013.


When integrating numerous in-car information and assistance systems, a consistent way of spatial distance presentation for drivers is required. A common practice is to use discrete textual information (e.g. 500 meters) in combination with a graphical bar representing relative spatial information. Hitherto there exists no design consistency with respect to bars for distance illustration. Contemporary solutions differ in terms of movement direction (upward vs. downward), composition type (decreasing vs. increasing), or alignment (horizontal vs. vertical). We conducted a driving simulator experiment to investigate user preferences, perceived location, and eye gaze data for a meaningful subset of bars in a dynamic scenario. When approaching a traffic event (road works), one out of four vertical bar alternatives indicated the current distance. Subsequently, the associated horizontal bar type (decreasing or increasing) visualized the driver’s spatial progress within the road works section. Our results indicate, that drivers prefer upwards-moving approach bars and progress bars decreasing from left to right. Eye-tracking data supports usage of decreasing bars instead of increasing bars. Accordingly, we elaborated an initial version of design guidelines for bars representing relative spatial information for local events. On this basis we implemented approach and progress indicators, which were adopted for numerous use-cases in a large field operational test for Vehicle-2-X Communication.


Weitere Links

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence