More and more connected and intelligent. The study on self-driving cars it is in full swing and the future of four-wheeled mobility promises cars capable of automatically picking up certain sounds and activating the correct maneuvers without the help of the driver.
Self-driving cars: the smart microphone
Cars are not yet equipped with a sense of hearing. In the future, however, systems capable of capturing and identifying external noises are expected to play a key role. Researchers from the Fraunhofer IDMT in Oldenburg are developing AI-based systems capable of recognizing individual acoustic events. These will give vehicles auditory ability.
The study conducted by researchers from the Fraunhofer Institute for Digital Media Technology focused on the behavior of motorists in certain situations.
For example, when an ambulance travels the siren roads often explained, those who are driving will panic and fail to perform the correct maneuvers to allow the transit of the rescue vehicle.
To prevent these "inconveniences" microphones have been created, similar in all respects to modern antennas dedicated to radio, through which to allow the car to hear in advance certain sounds.
These are connected to a control unit managed by a software developed specifically to allow theartificial intelligence of the car to make the correct decision by excluding the driver's intervention or by notifying the driver well in advance.
Self-driving cars: a smart "ear"
Thanks to the presence of microphones and specific software it is therefore possible to pick up certain types of sounds in advance so as to allow the driver to perform the correct maneuver without anxiety and stress.
The acoustic sensor system includes microphones, a control unit and software. The microphones, installed in a protective casing, are mounted outside the vehicle and capture airborne noises. The sensors transmit these audio data to a special control unit that converts them into the related metadata.
The Fraunhofer IDMT then designed its own algorithms beamforming. These allow the system to dynamically locate moving sound sources such as the siren on an approaching emergency vehicle. The result is one intelligent platform sensors capable of recognizing specific sounds.
The goal of the HSA, Speech and Audio Technology HSA Division of the Fraunhofer Institute for Digital Media Technology IDMT is to transpose the scientific results relating to hearing perception and human-machine interaction into technological applications.