A methodological approach to determine the benefits of external HMI during interactions between cyclists and automated vehicles: A bicycle simulator study

Abstract

To ensure safe interactions between automated vehicles and non-automated road users in mixed traffic environments, recent studies have focused on external human-machine interfaces (eHMI) as a communication interface of automated vehicles. Most studies focused on the research question which kind of eHMI can support this interaction. However, the fundamental question if an eHMI is useful to support interactions with automated vehicles has been largely neglected. The present study provides a methodological approach to examine potential benefits of eHMIs in supporting other road users during interactions with automated vehicles. In a bicycle simulator study, 20 participants encountered different interaction scenarios with an automated vehicle that either had the maneuver intention to brake or to continue driving. During dynamically evolving situations, we measured their behavior during interactions with and without eHMI. Additionally, the comprehensibility of the eHMI was measured with a special occlusion method. The results revealed that the eHMI led to more effective and efficient behavior of the cyclists when the automated vehicle braked. However, the eHMI provoked safety-critical behavior during three interactions when the vehicle continued driving. The set-up, experimental design, and behavioral and comprehension measurements can be evaluated as useful method to evaluate the benefits of any given eHMI.