MODELING OF DRIVER ANXIETY DURING SIGNAL CHANGE INTERVALS

The anxiety that a driver experiences at the onset of the yellow signal during the driver's approach to a signalized intersection is analyzed. The driver's decision is modeled as a reasoning process that consists of a set of fuzzy inference rules for stopping or continuing through the intersection. The input to the rules is the information on the current condition that the driver perceives. Because neither the rules nor the perceived information is clear, the driver's decision is associated with uncertainty. This uncertainty is quantified by possibility and necessity measures. Yager's anxiety measure is used to quantify the driver's anxiety associated with making decisions under uncertainty as a function of possibility and necessity measures for the conflicting actions. Anxiety is computed for both aggressive and conservative drivers. The measures for these two extreme types of driving behavior form the range; most drivers' behavior is believed to fall between the two. The model is used to estimate the degree of anxiety and its location on an actual intersection approach on the basis of the field data. The proposed method should be useful to evaluate the accuracy and the type of information to be provided to drivers and also to analyze the decision process of elderly drivers and drivers under the influence of alcohol and drugs.