Abstract
Keywords - driver performance, drowsiness, driver model, vehicle guidance, driver response characteristics
Aiming at increased road safety and a reduced number of accidents, research and development of motor vehicles focus more and more on the driver. This is due to the fact that, apart from the vehicle condition, the driver and his actual ability to drive a vehicle safely are essentially important for avoiding accidents.
The most recent research approaches for identifying the driver performance (DP) can be divided into two methods, the direct and the indirect method. In the first case, the method requires additional sensor systems, which monitor the driver or his actions directly. The indirect method includes all approaches that allow conclusions about the driver performance by acquiring parameters and their time changes, which specify the interaction between driver, road and vehicle (DRV parameter space).
In the scope of this article, two approaches and selected results of the indirect method are presented. The first approach is based on the analysis of how the driver keeps the vehicle in the lane, detected with a track monitoring system. By means of defined objective parameters, which characterise the driver-specific lane keeping behaviour, conclusions about the current driver performance can be drawn. The second approach is the identification from driver response characteristics. Conclusions about the current driver performance can be drawn by means of a driver model for lateral vehicle guidance and identified objective parameters. Based on a correlation and regression analysis, the objective parameters of both approaches are rated and appropriate objectification models, based on objective parameters, are generated.