The automotive industry in recent decades has been changing, electric vehicles are now increasingly common on the streets and each of the major car manufacturers has launched at least one model of electric or hybrid vehicle. This trend is likely to increase over time.
A characteristic of the electric motor is the low noise level at low speeds. At first glance, this would seem to be an advantage, as the noise emitted by vehicles contributes to increasing noise pollution, which has negative effects on people’s health and the environment.
However, the silence of an electric vehicle can be dangerous for pedestrians. The “clues” used by people to detect an approaching vehicle are divided into visual and auditory.
Using only visual clues, it has been found that pedestrians tend to underestimate low speeds and this can be dangerous. Furthermore, visually impaired pedestrians can rely almost exclusively on auditory clues, for these reasons the design of a system that increases the ease of detection of electric vehicles is even more important.
The solution currently used is to add an external sound to the vehicle in order to increase its detectability. This sound is called AVAS (Acoustic Vehicle Alerting System) and was regulated in October 2016 by the “Regulation n.138 of the Economic Commission for Europe of the United Nations (UNECE)” and by the “Regulation delegated by the EU Commission 2017/157” which established that from 1 July 2021 all-electric cars will be able to circulate only if they have an AVAS available which must have the characteristics indicated in the aforementioned regulation.
The aim of this project is to design a sound that complies with European legislation and to explore the different opportunities available during the sound design process.
Initially, the results of some researches already carried out on the subject were studied which analyzed the perceptual effect of some sound characteristics, in particular those that make AVAS more easily detectable. Furthermore, subjective parameters such as “appropriateness” and “pleasantness” were also considered.
Once these characteristics were defined, possible techniques for sound design were explored. The first and simpler technique consists in synthesizing a sound composed of several simple wave forms. As a starting point, we used one of the presets of the “Igniter” software, developed by “AIR Music Technology”, used as a basis on which to work. As a result, it was possible to add an indefinite number of components using the main waveforms (sinusoidal, square, sawtooth, triangular) or possibly a middle ground between two of these basic waveforms. The other technique used is that of granular synthesis, which consists in using recorded sound samples and reproducing only small pieces in a loop, acting on different parameters and introducing a factor of randomness. This technique will be explored in detail and is the one most used by the various car manufacturers, as it allows a more creative approach and different results. After all, the need to add an AVAS to the vehicle is also an opportunity to brand the product, generating innovative sounds that can be easily associated with a car manufacturer
Design of a sound driven by an advanced system to sound electrified vehicles (AVAS) in compliance with current regulations.
We wanted to create a sound that in the most efficient and effective way possible returns, upon listening, the sensations that a pedestrian perceives when a vehicle approaches.
Sound design via SuperCollider and granular synthesis. Validation of the system in an anechoic chamber to verify compliance with the regulations in terms of sound power and audio frequencies.
The approach of sound generation through granular synthesis is particularly suitable for the design of sounds through a mix of elementary acoustic elements called microsounds or grains. These elements, suitably combined and modulated, can be exploited to generate more articulated sounds and with a dynamic behavior dependent on data from external sensors.
Analysis of the porting of the activity on an embedded DSP board with less resources and lower costs. Analysis of the application of the model to sectors other than Automotive (e.g. railway, industrial, …)