Adaptive in-car user interfaces based on personalized workload estimation

Based on recent research on workload scheduling and personalization, we developed a personal electronic driver assistant that mediates the interactions between the driver and in-car services in order to prevent overload. Whereas other approaches for overload prevention often focus on specific workload sources (e.g. phone calls), our assistant takes into account a broad range of factors: individual differences (e.g. experience), the general driving context (e.g. road condition) and the overall information supply in the car (e.g. internet). As an addition to current approaches, we developed a prototype assistant that takes into account driving experience, age and presence of passengers in the car. Subsequently, we conducted a first experiment to test the effects of driver’s experience. Unfortunately this experiment did not provide clear results on the costs and benefits of the support. We did identify high expectations of a personalized driver’s assistant, which current systems apparently cannot meet. Keywords: Smart transport, in-car services, workload, adaptive user interface, personalization.