What variables affect to a greater extent the driver's vision while driving?

Abstract

This paper deals with the analysis of the main variables involved in the visual activity of a driver of motor vehicles, in order to identify the most important quantities and implement, therefore, appropriate corrective actions to the achievement of road safety. The first step in this research was to survey a number of variables within the road environment and processing this data base with clustering techniques in order to extract useful information for purpose. In this case, a mixture of procedures based on Fuzzy Logic (FL) and Artificial Neural Network (ANN) were applied not only to extract knowledge not known a priori but, above all, to define the membership functions and rules of the fuzzy model without recourse to the skills of the analyst, not always so objective. This procedure, applied to a rural road open to traffic, showed a good performance in predicting the user's visual behavior and, especially, in identifying the most influential variables. This aspect may allow the agency to direct the maintenance operations so that to facilitate understanding of the information contained within the road environment, thus improving safety. © 2013 Copyright © 2013 Vilnius Gediminas Technical University (VGTU) Press.