Drivers’ behaviour when overtaking cyclists on rural roads: Driving simulator validation using naturalistic data

Abstract

Driving simulator is a useful tool for obtaining data on driver behaviour efficiently and quickly. However, to ensure the reliability of the data gathered by the simulator, it is necessary to check the differences between drivers’ behaviour in the simulator and in reality. In this study, an existing two-lane rural road was replicated on the driving simulator under the same traffic conditions of groups of cyclists and oncoming motorised vehicles. For this purpose, a naturalistic field data collection was developed on the real road using instrumented bicycles and static video recordings. A total of 30 volunteers participated in the driving simulator tests. The objective validation of the driving simulator was based on three operational variables: average travel speed, overtaking vehicle speed, and lateral clearance. As a result, higher average travel speeds and lower lateral clearances were obtained in the real world compared to those observed in the simulator. It was also found that overtaking vehicle speed depends on the group of cyclists. Overall, the data obtained in the field and in the driving simulator did not present statistically significant differences. The analysis of drivers’ perception in the simulator tests concluded that the simulator reflected reality in an accurate way, achieving the subjective validation of the driving simulator. Thus, this study validates the driving simulator for bicycle safety research on rural roads.