|Bibliography||Heinrich, Frederik: Vorhersage der Fahrerbelastung während der Fahrt. |
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Diploma Thesis No. 3311 (2012).
114 pages, german.
As interactive systems in cars are on the rise, driver distraction emerges as an important issue for the automotive industry. When car drivers operate different devices while driving, less attention is attributed to the primary task of driving safely. Therefore, various systems are being created to improve driving security. One of them is an adaptive system that predicts the driver’s mental workload. Data from various origins can be taken into account to estimate the workload. There are physiological sensors (e.g. heart rate, skin temperature), car sensors (e.g. steering angle, rain sensors), data from the environment (e.g. traffic, weather) and real-time data on the condition of the road’s surface.
This thesis presents a conceptual system for workload estimation based on a selection of various parameters. A field study with ten test persons was conducted to find out whether different driving environments (e.g. highway, inner-city roads, roundabouts) produce a measurably varying mental workload. For the measurements, various physiological data was collected. In addition, a video-rating served as a subjective measure for later comparison with the physiological data. According to the evaluated field data, the driver’s skin conductance, known as a reliable indicator for mental workload, correlates the most with the results of the video rating. Speed-limited roads (30 km/h) and roundabouts proved to be the driving environments provoking the highest mental workload. Based on this knowledge, a paper-prototype system was developed which takes into account the mental workload scores of the field test data. Several other data sources such as weather and traffic information were added as supplementary parameters to predict the driver’s mental workload. Finally, two cases are presented as examples of applying mental workload estimates to increase driving safety. In the first case, a prototype application adjusts the information density on the GPS navigation screen according to the predicted mental workload. The second case describes an adaptive communication system which restricts or blocks means of communication (e.g. incoming telephone calls) based on the driver’s estimated workload. By making use of such adaptive systems, driver distraction can be reduced leading to increased driving safety.