Electric vehicles, being inherently quiet without the typical combustion noises, pose a potential safety concern, especially at low speeds. Consequently, an Acoustic Vehicle Alerting System (AVAS) is mandatory in many countries worldwide to warn pedestrians of approaching electric vehicles. The development of AVAS sounds involves conducting measurements on an outside noise test track to verify compliance with regulations. Various environmental parameters on the test track can influence the transmission of sound from the car’s AVAS speaker to the measurement microphones.
This research delves into understanding the relationship between the transmission of sound over short distances and environmental parameters. Over a one-year period, 122 measurements were conducted using a specially designed dolly setup. The frequency response function, which characterises the sound transmission, was calculated to determine the dependencies and correlations with environmental parameters.
The findings reveal a significant dependency on environmental parameters, as indicated by a linear regression model. Frequency-based correlation analysis indicates variations in dependency across the spectrum. Furthermore, the analysis suggests that sound transmission is more effective in denser mediums, such as colder air temperatures and higher air pressure. The density of the medium, derived from the measured parameters, exhibits a high dependency.
These results now enable predictions of sound transmission on the outside noise test track under user-defined environmental conditions using artificial intelligence. This advancement not only enhances our comprehension of variations in test track results but also facilitates the prediction of future outcomes.