Driver behavioural adaptations to simulated automated vehicles, potential implications for traffic microsimulation

Zwart, R. de; Kamphuis, K.; Cleij, D.

The ongoing technological development of automated vehicles is bringing their implementation ever closer. Until all vehicles are completely automated there will be a mix of human driven and automated vehicles. To assess impacts of such automated vehicles on traffic safety and flow, techniques involving traffic microsimulations are often used. Here mixed traffic scenario’s are simulated by assigning different behavioural models to automated and to human driven vehicles. The human driver models generally remain unchanged when simulating different penetration rates of automated vehicles. In the real world, however, phenomena like social contagion could result in changes in human driving behaviour when automated vehicle penetration rates increase. To investigate such adaptive behaviour a simulator study was conducted where participants were asked to drive the same route three times, each time with a different penetration rate of automated vehicles. The simulated automated vehicles were assigned driving behaviour that resulted in a shorter time-headway, stricter speed control and a reduced reaction time compared to the simulated human drivers. The results show that when the penetration rate of automated vehicles increased the participants adopted shorter time-headways, smaller following distances, less variable velocity more closely resembling the maximum allowed speed, and reduced reaction times with faster acceleration at traffic lights. These findings suggest that driving behaviour changes when the composition of surrounding traffic changes. In order to more accurately use traffic microsimulation to estimate effects of mixed traffic it is therefore important to ensure that human driver models are updated to incorporate social contagion effects.

Published in
Transportation Research Part F: Traffic Psychology and Behaviour
92 (January 2023)

SWOV publication

This is a publication by SWOV, or that SWOV has contributed to.