Improved road safety as AI teach cars how to hear

Infineon Technologies AG, together with Reality AI, produced an advanced sensing solution providing vehicles with the sense of hearing. This solution adds XENSIV MEMS microphones to existing sensor systems. It allows cars to 'see' around the corner and to warn about moving objects concealed in the blind spot or approaching emergency vehicles that are still too far away to see.

The new sensing solution is based on XENSIV MEMS microphones together with AURIX MCU and Reality AI’s Automotive See-With-Sound system. Employing machine learning-based algorithms, the system can detect emergency vehicles, cars and other road participants – even if they are unseen by drivers or detected by the sensors incorporated in the vehicles’ ADAS. Machine learning also ensures that the country-specific sirens of emergency vehicles are recognised in all parts of the world.

The automotive-qualified XENSIV MEMS microphone IM67D130A provides an increased operating temperature range from -40C to +105C enabling various use cases in severe automotive environments. The low distortions and the high acoustic overload point of 130dB SPL allow the microphone to capture distortion-free audio signals in noisy environments. This allows a reliable classification, even if the siren sound is hidden in the high background or wind noise. This sound-based sensing technology can also facilitate other applications in vehicles such as road condition monitoring, damage detection or even predictive maintenance.

For processing the audio signal, the Reality AI software uses Infineon’s AURIX TC3x family of MCUs. The scalable MCU family provides a range from one to six cores and up to 16MB of Flash with functional safety up to ASIL-D according to the ISO26262 2018 standard and EVITA full cybersecurity. AURIX TC3x offers customers the performance and flexibility to implement XENSIV MEMS microphones in ADAS use-cases.