Study of the effectiveness of electric vehicle warning sounds depending on the urban environment

Abstract

Electric and Hybrid Electric vehicles (EV and HEV) seem to be the future of transport in smart cities and the key for the total reduction of noise disturbance and pollution in urban areas. However, several problems have to be solved to guarantee the safety of these types of vehicle. Up to now, the use of HEV has shown the dangers of a “quiet” transport system in urban environments; in fact, it has been estimated that an HEV is twice as likely to be involved in a pedestrian crash as would be an internal combustion engine (ICE) vehicle in the same situation. With the aim of improving their safety, different kinds of warning sounds are being designed to increase the detectability of EV and HEV without themselves becoming new annoying sound sources. The sound directivity, frequency of emission and intensity of warning sound systems will guarantee their efficiency and limited noise impact. Several research works bring to the fore a significant variation of the pedestrian response time to different acoustic stimuli. However, it is necessary to examine the suitability of these warning sounds according to the urban environments in which they are going to be emitted. Distinct areas inside the city have different soundscapes whose spectral content can vary significantly, masking some of the sounds suggested as an alert. This paper analyses in detail the main characteristics of several warning sounds proposed by the industry, conducting a comparative study of the different design trends. A total of 131 sighted listeners were exposed to a virtual road-crossing test. The behaviour and appropriateness of warning sounds are analysed depending on the urban environment. For this purpose, three clearly different soundscapes have been selected: stopped vehicles at a traffic light, a pedestrian shopping area and the vicinity of a playground. The results highlight the wide variability in pedestrian reaction time for the different warning sounds used. Some signals considerably improve the detectability of the vehicle, providing results even above the ICE vehicle ones. However, other warning sounds do not decrease the reaction time with respect to the EV. In addition, a clear dependence is observed between the detectability and the soundscape involved, changing the results for the same warning sound depending on the acoustic environment.