[Image: Vehcile Learns to Hear]

Vehicle Learns to Hear

Driver assistance systems and automated driving functions must be able to recognize the vehicle environment and the objects located in it. Up to now, camera, radar and lidar sensors have mainly been used for this purpose. As a new sensor concept, fka, together with its scientific cooperation partner, the Institut for Automotive Engineering (ika) of RWTH Aachen University, has been researching the development of automotive-suitable sound sensors and functions based on them under the name Sonic Intelligence for over two years. These are intended to give the vehicles their own sense of hearing.

When developing new functions for Sonic Intelligence, the fka experts for automated driving contribute basic methodological knowledge and expertise in the development of new sensor concepts, sensor data processing and fusion, function development and function validation and verification.

With this extensive experience, several functions based on sound sensors have already been successfully implemented. Among other things, a test vehicle is able to detect and locate the acoustic signals of emergency vehicles and accident events. In a prototypical overall system, the test vehicle was able to issue recommendations for actions to the driver via a specifically developed HMI. In addition, diagnostic functions have already been developed and implemented that support the system and determine the reliability of the prediction, for example by detecting sensor errors in the event of sensor contamination. A decisive comfort function has already been implemented with voice control. The next goal for Sonic Intelligence is a further integration, which also requires the fusion with other sensor data.

Machine learning methods are an important component in the implementation of the detection algorithms of the Sonic Intelligence technology. Here, fka can draw on extensive expert knowledge in the field of artificial intelligence, particularly in the area of machine perception. In addition to the know-how about suitable network structures and the necessary data preprocessing, the availability of a large database with reference information is also a requirement for the successful use of these procedures. In recent years, fka has already built up a corresponding database and process chain, which enable rapid development.

Are you looking for further information or are you interested in a cooperation? Please feel free to contact us at any time regarding this topic. We look forward to receiving your message!

[Foto: Vehicle Learns to Hear]

Testing infrastructure

Ihr Ansprechpartner

Your contact

[Foto: Adrian Zlocki]

Bereichsleiter Automatisiertes Fahren
Dr.-Ing. Adrian Zlocki

Manager Business Unit Automated Driving
Dr.-Ing. Adrian Zlocki

This email address is being protected from spambots. You need JavaScript enabled to view it.
Telefon: +49 241 8861 219

This email address is being protected from spambots. You need JavaScript enabled to view it.
Phone: +49 241 8861 219

fka World of Research

[Foto: PV1000 -The People's Ventilator]
Voluntary Commitment | Julian Refghi M.Sc.

PV1000 - The People's Ventilator

A team of volunteers is developing the PV1000, a simple ventilator for intensive care that can be used as soon as the stock of conventional ventilators becomes scarce.

Read more ...

[Foto: fka Digital Events]
Corporate Development | Dr.-Ing. Markus Bröckerhoff

fka Digital Events

We focus on digital events and content – not only now, but also in the future.

Read more ...

[Foto: Sound.AI] Highlight-Project | Thomas Keutgens

Sonic Intelligence

fka experts from the fields of acoustics, electrics and electronics, strategy and consulting as well as automated driving, together with experts from ZF, have made the listening vehicle real.

Read more ...

read more news