Effect of roadside configuration and cross-section on driver’s behaviour

This paper describes the results of a driving simulator study focused on the influence of the roadside configuration and cross-section on driver’s behavior. The research was prompted by the consideration of the high accident rate on two-lane rural roads lined with trees. This is a layout currently adopted on many European rural roads. An experiment was designed in order to: a) study the effect of roadside elements on driver’s behavior; b) evaluate the influence of roadside elements also depending on different geometric features. A two-lane rural road lined with trees was designed with the alignment including a wide range of geometric features. Two different cross-sections (with and without shoulder) combined with three lateral configurations were tested. Thus, six road scenarios were obtained. Thirty-six drivers drove in the simulator on these scenarios and the values of speed and lateral placement were collected. The statistical analysis showed, for both tangents and curve segments, that drivers did not behave differently in terms of lateral placement and speed varying the roadside configuration but a significant difference was found only for the shoulder factor. This fact suggests that the presence of trees without barrier is not perceived as a higher risk and thus, it should be avoided. The study also revealed that drivers tend to move away from the right side of the road at around the point where the guardrail is introduced. This effect was significant for both left and right curvatures (particularly in case of absence of the shoulder) but not for tangents.