Abstract
It is well understood that a significant shift away from fossil fuel based transportation is necessary to limit the impacts of the climate crisis. Electric micromobility modes, such as electric scooters and electric bikes, have the potential to offer a lower-emission alternative to journeys made with internal combustion engine vehicles, and such modes of transport are becoming increasingly commonplace on our streets. Although offering advantages such as reduced air pollution and greater personal mobility, the widespread approval and uptake of electric micromobility is not without its challenges. Concerns have been raised regarding the safety of such vehicles, most notably related to pedestrian safety of blind and partially sighted individuals, due to the inherently lower sound levels produced by electric vehicles. This study addresses this issue by investigating the use of an Acoustic Vehicle Alerting System (AVAS) for electric scooters by means of a virtual reality experiment and field trials. Eighty-eight participants from four European countries, including thirty-five blind or partially sighted individuals participated across the experiments. Results show high missed detection rates for electric scooter operations without an AVAS in typical city soundscapes (90-97%) and an increase in detectability for all AVAS conditions tested. Modifying AVAS sounds with playback rate and level changes with respect to operational state facilitates detection of deceleration, as well as improving detectability in multiple vehicle scenarios.