NAOMI4Radar: More Energy Efficiency and Safety for Autonomous Driving through Neuromorphic Computing

Energy efficiency and safety are central requirements in autonomous driving. As part of the "NAOMI4Radar" project, we are developing innovative solutions in collaboration with Mercedes-Benz, the University of Lübeck, and associated partners Intel Labs and the Technical University of Munich through the use of neuromorphic computing.

The goal of "NAOMI4Radar" is to optimize radar data processing in autonomous vehicles using #NeuromorphicComputing. Compared to traditional AI algorithms, this approach offers significant advantages: lower energy consumption reduces CO2 emissions, while direct processing of algorithms on the chip decreases latency, thereby increasing safety through faster response times. Within the approximately one-year project duration, we aim to demonstrate the industrial applicability of the neuromorphic chip #Loihi2.

The project is practically and serially oriented to enable a direct transfer of results to the automotive industry. TWT contributes to the project realization by developing a prototype demonstrator, which integrates the entire processing chain in an end-to-end manner—from data acquisition with radar sensors, processing via the neuromorphic chip, to the visualization of results in the vehicle.

We are excited to bring our expertise from numerous (inter)national research projects in the fields of #AI, #NeuromorphicComputing, #ADAS, and #Sustainability to this pioneering project.

The "NAOMI4Radar" project is publicly funded by the #BMWK and managed by the project sponsor TÜV Rheinland.