Accurate environmental perception is crucial for automated driving, with automotive radar sensors being particularly essential in adverse weather conditions, such as rain and fog. Simulation has emerged as a valuable tool for minimizing the validation efforts required for testing driving functions and sensors in real-world environments. Critical scenarios and weather conditions can be reliably and independently simulated using virtual environments. However, to ensure the reliability of these simulations, it is necessary to validate the simulation tools used in this study by comparing them with real-world scenarios. This thesis investigates radar signal intensity without accounting for the angle of arrival and Doppler variations, focusing solely on actual radar sensors without integrating them into real-time measurement systems. The thesis demonstrates how radar signals weaken when detecting drops in the immediate vicinity, thereby providing insights into how precipitation influences radar performance. It highlights the significant impact of radome coverage with salty water from asphalt, thereby simulating conditions in which road spray affects sensor accuracy. By examining material properties, this research enhances the understanding of radar behavior in complex urban environments. Further, this research demonstrates how different materials can affect radar signals and detections, thus questioning the belief that radar is a robust sensor. This thesis reveals that heavy rain and spray from vehicles can notably hinder radar performance, thereby reducing its ability to accurately classify objects. These results raise concerns regarding the dependability of radar sensors in all weather conditions and should be considered and modeled for further automated driving developments.
Diogo Wachtel Granado (Thu,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: