Electric Vehicles (EV) are characterized by a high reduction of the acoustic emission. The absence of warning sounds entails a risk situation for pedestrians. The previous research is focused on detectability of warning sounds in different noise environments. These experiments are performed indoors, where a pedestrian’s conditions are not similar to real road crossing. Drivers’ behaviour study demonstrated that different environments and workload have influence on reaction time. Consequently, this paper proposes a methodology for the analysis of detectability of real warning sound using a dynamic subject. The sample was composed by 65 participants walking around a pedestrian area. Participants had to react when they detected a vehicle approaching. The subject’s response was affected by background noise, therefore, this parameter was measured. The results establish that power levels have influence on the detectability. There is an optimum power level which improves efficiency of vehicle detection. Besides, warning sound features and learning effect, based on previous experience, have influence on subject response. (Author/publisher)
Samenvatting