From ML/AI Target Classification to 512-Channel 4D Radar: Ansys HFSS SBR+ for Simulating Radar Sensor Models

"Advanced driver assistance systems (ADAS) have revolutionized the automotive industry by significantly improving vehicle safety and providing a path to realizing fully autonomous vehicles. Level-1 and Level-2 automation have already been deployed in commercially available vehicles providing features such as parking, steering, traffic-jam, and lane control assistance. A crucial requirement for safety related ADAS is constant and reliable situational awareness in a highly dynamic environment around the vehicle.

Radar has emerged as a core sensing technology for many active-safety and comfort related advanced driver assistance systems (ADAS) being deployed in today's vehicles. While measurement is valuable, it can be costly, time consuming, restrictive due to practicality issues and also unsafe in some corner-case traffic scenarios. Simulation has emerged as an attractive alternative source of synthetic radar returns for radar sensor development.

Ansys has emerged as the industry leader in radar simulation with the HFSS SBR+ solver. Using HFSS SBR+, customers can obtain highly accurate, high-fidelity physics-based radar returns that can be used for designing, testing and validating radar sensors. Using groundbreaking innovations in HFSS SBR+, we show how simulation has been used to obtain range, Doppler, azimuth and elevation angle-of-arrival in full-scale traffic scenes using a 512-channel 4D radar sensor. Metallic guardrail posts have been known to present unique corner cases for radar operation. Here, we present the most comprehensive treatment of the guardrail corner case scenario using radar cross section simulations, time domain animations and range-Doppler maps. We further demonstrate how Ansys technology can be used to obtain vast quantities of data required for training ML/AI algorithms to identify radar targets using Microdoppler. Over 7 peer-reviewed journal articles have been published by Ansys in relation to these topics.

"