The report consists of two parts. The first part is a literature study aimed at identifying the most relevant indicators for predicting and detecting fatigue induced impaired driving. The aim was to identify possible performance based technologies, such as the lateral control of the car. Thus, less attention was devoted to e.g. psycho-physiological measures like EEG. A large number of research reports describing various approaches have been reviewed. The survey clearly indicates that no single indicator can be used to detect drowsy driving. A combination of different measures is recommended e.g. analysis of lateral control performance and eye blink pattern. Furthermore, it should be noted that so far there is no commercial system available that provides a sufficiently reliable method to detect a drowsy driver. In the second part, experimental data from previous driving simulator experiments were analysed by means of signal processing and statistical analyses. Specific attention was paid to investigate the potential of lateral vehicle position data as a means to estimate driver drowsiness. The analysis of experimental data did not reveal any clear answer to what indicators are the most prominent with respect to detect drowsy driving behaviour. Further investigations and analyses of driving behaviour data are needed. However, the data analysis conforms in large to the findings in the literature survey. (Author/publisher)
Abstract