Evaluation methodology for physical radar perception sensor models based on on-road measurements for testing and validation of automated driving

In recent years increasingly verification and validation processes of automated driving system have been moved to virtual simulation, as it allows rapid prototyping and using a multitude of testing scenarios compared to on-road testing. However, in order to support future approval procedures for automated driving functions with virtual simulations, the models used for this purpose must be sufficiently accurate to be able to test the driving function implemented in the complete vehicle model. In recent years, the modeling of the environment sensor technology has gained particular interest, since it can be used to validate the object detection and fusion algorithms in the model-in-the-loop (MIL). In this paper a practical process is developed which enables a systematic evaluation for perception sensor models on a low-level data basis. The validation framework includes the creation of test maneuvers, the execution of measurement runs on a closed motorway section of the M86 in Hungary, where all dynamic objects are equipped with a reference measurement system, the setup of a precise digital twin of this highway section, re-simulation of test maneuvers using low-level sensor models, the automatic process for labelling sensor output to the ground truth objects and the evaluation of the sensor model according to statistical metrics.To demonstrate the practical feasibility, a commercial RADAR sensor model (ray-tracing RSI RADAR model from IPG) was validated using a real RADAR sensor (ARS-308 RADAR sensor from Continental). The simulation was set up in the simulation environment IPG CarMaker® 8.1.1 and the evaluation was then carried out using the software package Mathworks Matlab®. Real and virtual sensors output data on a low-level data basis was used, which thus enables the benchmark. Metrics are developed for the evaluation, which are quantified using statistical analysis.