Frangipani
Top 20
Speaking of NAOMI4Radar: Presentations on the project’s results have begun… Like this talk earlier today by Franziska Ott (TWT GmbH Science & Innovation) and Geoffrey Kasenbacher (Mercedes-Benz) at ELIV (Electronics in Vehicles) 2025 in Bonn:
#eliv2025 #naomi4radar #mercedesbenz #twtgmbh #radartechnology #autonomousdriving #neuromorphiccomputing #adas #innovation #collaboration | Dr.-Ing. Johanna Kasper
🎤 On stage at ELIV 2025 – Naomi4Radar: Advancing Autonomous Driving with High-Efficiency Neuromorphic Radar Heute auf der Bühne bei der ELIV 2025: Franziska Ott (TWT GmbH Science & Innovation ) und Geoffrey Kasenbacher (Mercedes-Benz AG ) präsentierten im Track “Automotive AI – From Sensor to...www.linkedin.com
View attachment 92032
English translation:
View attachment 92033
On this slide we can see which specific researchers from each of the project partners were involved in a publication resulting from the project…
View attachment 92034
And here you can see Loihi 2 (the blue box on the left) as well as the Infineon front radar used…
View attachment 92035
In case any of my fellow German speakers would like to check out the (pretty technical) NAOMI4Radar project final reports that the three consortium members - Mercedes-Benz, TWT GmbH Science & Innovation and Uni Lübeck - were obliged to submit due to having received public funding for their project: the “Schlussberichte” have now been uploaded to RENATE (Repositorium für Naturwissenschaften und Technik), which also “serves as the repository for the publication of reports to federally funded research projects”.
While there is both a German and an English version of the accompanying one-page document below, all the final reports are in German only. Each of the three consortium partners submitted an individual report to what was formerly known as Germany’s Federal Ministry for Economic Affairs and Climate Action that had funded 56% of the total 900,000 € project costs (cf. page 15 of https://www.bundeswirtschaftsminist...projektsteckbrief-automatisiertes-fahren.pdf?)
In addition, a more extensive joint report by all three consortium members was also submitted - that’s the one, which I have linked below. Note that while they collaborated with both Intel (who provided the physical Loihi 2 hardware as well as technical support) and Infineon (who provided the front radar and assisted with its integration), those two companies were not consortium members themselves and didn’t receive any of the public funding. Neither was TU München/Munich, whom the consortium also collaborated with.
By the end of their 15 month long project (1 June 2024 - 31 August 2025) titled NAOMI4Radar (Neuromorphic Algorithms for Optimizing Automotive Radar Sensor Processing in Autonomous Vehicles), the researchers had reached TRL* 7 as planned, which is defined as “System prototype demonstration in operational environment”. At the start of the project, their research was considered as being TRL 3.
*TRL = Technology Readiness Level - the highest on the scale is TRL 9.
By the way, Akida gets a mention, too, but not - as one would expect - in the context of the Vision EQXX (page 8):
German original:
“Neuromorphic Computing
Bereits im Konzeptfahrzeug Vision EQXX wurde die Leistungsfähigkeit neuromorpher Ansätze zur Verbesserung der Energieeffizienz demonstriert. Um die Technologie weiterzuentwickeln, arbeitet Mercedes-Benz eng mit Startups, Forschungseinrichtungen und Studierenden zusammen. Dies unterstreicht die Bedeutung von Neuromorphic Computing für das Unternehmen und dessen zukünftige Fahrzeugarchitekturen.”
Translation into English (done by DeepL & refined by Frangipani):
“Neuromorphic computing
The capability of neuromorphic approaches to improve energy efficiency has already been demonstrated in the Vision EQXX concept car. Mercedes-Benz is working closely with start-ups, research institutions and university students to further develop the technology. This emphasises the importance of neuromorphic computing for the company and its future vehicle architectures.”
Instead, BrainChip and Akida get mentioned towards the end (page 35), in a paragraph describing other entities’ progress in this field of research (= radar data processing) that was revealed during the duration of the NAOMI4Radar project.
For the German original, see below.
English translation:
“During the course of the project, it became known that in the US BrainChip Holdings, together with Raytheon, had been awarded a contract for neuromorphic radar processing by the Air Force Research Laboratory. The aim of the project is to implement complex signal processing algorithms, in particular micro-Doppler signature analysis, on neuromorphic hardware. BrainChip's Akida processors offer ultra-low energy consumption and are suitable for high-performance signal processing on energy- and space-constrained platforms such as drones or other mobile systems. The project underlines the growing interest in neuromorphic technology for real-time applications in radar and AI in both the military and commercial sectors.”
NAOMI4Radar - Neuromorphe Algorithmen zur Optimierung der Radar-Sensorik in autonomen Fahrzeugen
Das automatisierte Fahren ist ein für die deutsche Automobilindustrie strategisch wichtiges Innovationsthema. Bei der Entwicklung und dem Einsatz entsprechender Technologien hat die deutsche In-dustrie in den vergangenen Jahren große Fortschritte gemacht. So kommt z.B. die weltweit erste...
[…]
[…]
