Individual driver safety awareness assessment using simulating experiment data

Abstract

Driver safety awareness level has important influence on traffic safety, in order to extract and analyze the factors and events which have a significant effect on driver safety awareness while driving. This research uses virtual reality technology to design and implement several kinds of typical virtual traffic scenarios for safety consciousness level evaluation, and sets the assessment scenarios of intersection as an example, carries out 80 groups of driving simulation experiments, collects several experimental data including driver status, vehicle status and traffic environment, etc. On this basis, logistic regression model is used to consider the three factors, namely human-vehicle - road, and establish the driving safety awareness level evaluation model under the typical traffic scene. It is found that the driver experience has important influence on its security level of consciousness. The results show that the area under the ROC curve is 0.918, which shows that the established bi-level logistic model in terms of driver safety awareness evaluation and forecast has good adaptability and practicability.