Tour de Interaction: Understanding Cyclist-Driver Interaction with Self-Reported Cyclist Behaviour

How Autonomous Vehicles (AVs) and cyclists will interact in shared traffic is largely unexplored but is key to ensuring safety when these new vehicles join our roads. We report the results of an online survey (n=383) designed to understand current cyclist-driver interactions as a starting point to inform how AVs should behave around riders. We asked cyclists to identify cyclist-driver interaction techniques and scenarios and self-report their understanding of AVs. We found that cyclists rely mainly on eye contact and hand gestures to exchange awareness, intent and appreciation with drivers. Therefore, AVs must recognise and respond appropriately to such cues. We also found that cyclists have a higher perceived risk in scenarios with little traffic control, such as lane merging. Riders did not know enough about AVs to reach a consensus concerning their impact on cyclist behaviour. We conclude by discussing the implications of our findings on AV interaction design.