AuRaSim: A Ray-Tracing-Based Automotive Radar Simulator for Dynamic Digital-Twin Scenarios

Conference Name
1 HELLA GmbH & Co. KGaA, Lippstadt, Germany
2 Dortmund University of Applied Sciences and Arts, Dortmund, Germany
3 Brandenburg University of Technology, Cottbus, Germany

Abstract

This is a template you can use to create a professional-looking, accessible, and performant project page for your research paper. See the code for the template and instructions on how to use it yourself on GitHub. It’s made with the Astro web framework and styled with Tailwind. You write the content in MDX, which enables markdown formatting like bold, italics, and strikethrough, as well as custom components like small caps.

Demonstrations

Digital Twin Scenarios

Real scene for digital twin comparison 1
Simulated digital twin scene 1
Real scene for digital twin comparison 2
Simulated digital twin scene 2
Real scene for digital twin comparison 3
Simulated digital twin scene 3
Real scene for digital twin comparison 4
Simulated digital twin scene 4

Ray-Tracing

Ray-Tracing Demo 1

Double-click to focus · Wheel to dolly

Ray-Tracing Demo 2

Double-click to focus · Wheel to dolly

Video

Parking Lot — Pedestrian Approaching

Scenario

Range-Doppler Map

Micro-Doppler

Parking Lot — Car Approaching

Scenario

Range-Doppler Map

Micro-Doppler

Sim2Real

RAMP-CNN Digital Twin

Camera

BMS1000

Real Radar

BMS1000

Sim Radar

BMS1000

Camera

PMS2000

Real Radar

PMS2000

Sim Radar

PMS2000

Quantitative Comparison

Quantitative comparison between measured and simulated Range–Doppler maps in terms of RMSE and PSNR. ↓ denotes lower is better and ↑ denotes higher is better.

ScenarioMethodRMSE ↓PSNR [dB] ↑
BMS1000RadarSimPy0.29811.94
AuRaSim0.17215.27
PMS2000RadarSimPy0.18915.46
AuRaSim0.15916.11
MLMS001RadarSimPy0.23312.90
AuRaSim0.11119.40
AverageRadarSimPy0.24013.43
AuRaSim0.14716.93

Bibliography

  1. Blender Foundation, Blender. Open-source 3D creation suite. Accessed Oct. 1, 2026.
  2. Fayçal Aït Aoudia, Jakob Hoydis, Merlin Nimier-David, Sebastian Cammerer, and Alexander Keller, Sionna RT: Technical Report, arXiv:2504.21719, 2025.