Parking assistants are often included as standard equipment in modern cars − convenient, smart, and seemingly flawless. But what happens when the technology fails? A scratch on the paintwork is harmless, but an incorrectly calculated angle can quickly lead to expensive damage or even pose a danger to pedestrians. The truth is: convenience alone is not enough. These systems must perform reliably under all conditions. That is where the real challenge emerges: How do you test a function that must account for countless variables in the complexity of the real world? Traditional methods are reaching their limits. One answer is vehicle-in-the-loop (VIL) testing − a validation method that combines reality and simulation, setting new standards. UN Regulations such as UN ECE R171 (“DCAS”) even propose using simulation for virtual homologation. Time to take a closer look.

End-to-End Validation Across All Test Levels: SIL, HIL, and VIL

Best practice is a combination of simulation methods such as SIL & HIL testing with real-world testing. Introducing the vehicle-in-the-loop (VIL) testing method enables road testing to be accelerated by reusing tools from SIL and HIL simulations on the actual vehicle.

To validate automated parking assist (APA), ultrasonic sensors are stimulated on a real vehicle in dynamic driving mode. During slow driving, e.g., during parking maneuvers, objects are defined in any virtual scenario, and object distances are calculated at run time in the simulation environment. These distances are the target values for the ultrasonic stimulators of the ultrasonic sensors installed on the vehicle. To validate the parking system's performance, the echo signal can be generated in real time. In addition to the time of flight, parameters such as amplitude, frequency, and number of signal oscillations can be adjusted. 

Consequently, simulated environmental data can be varied as desired and validated on the actual vehicle. Critical environments, such as parking garages with tight parking spaces, poles, and other types of barriers, can be simulated again and again. Variations in surfaces reflectivity can also be included in the tests. 

Finally, feedback to the vehicle’s real human-machine interface (HMI) can also be validated based on the simulated environment.
 

Vehicle-in-the-Loop Requirements and Setup

The most important requirements for the validation of parking assistants in a VIL test system

• Physically correct models of the vehicles, sensors, and environment.
• Simulation of typical parking scenarios like narrow gaps, curbs, and critical scenarios like highly reflective surfaces or unusual obstacles.
• Minimal latency between virtual environment and hardware.
• Precise synchronization between vehicle inertial measurement unit (IMU) and simulation.
• Closed-loop operation for dynamic scenarios.
• Error injection (e.g., noise, occluded objects) for robustness testing.
• Real-time analysis of fusion results.
• Easy parameter variation to generate a big variety of test cases.
• Logging and replay function for error analysis.
   

VIL Setup and Components

Device under test (DUT)

The DUT is the automatic parking assistance system installed on a real vehicle – a vehicle prepared for over-the-air-testing. 

Vehicle positioning and motion sensors (GNSS & IMU)

In this setup, the device under test (DUT)—in this case, the automatic parking assistance system—makes decisions on steering, braking, or throttle commands based on its perception of the environment. The key is to measure vehicle position, orientation, and motion accurately in real time. Without accurate real-time localization, the simulated environment can fall out of sync with the actual vehicle, resulting in incorrect stimuli being sent to the DUT.

Physics-based simulation for ultrasonic sensors

The ASM environment and ASM traffic models are calculated on a real-time PC with access to the automotive bus. Standards such as ASAM OpenDRIVE® can be used to convert a real network into a format that can be used by the simulation system. The physics-based sensor simulation AURELION creates a 3D representation of the ego-vehicle, its environment, and surrounding traffic participants. While AURELION can generate raw data for all environment sensor types, its capabilities also extend to the generation of raw ultrasonic sensor data. For ultrasonic simulation, AURELION models wave propagation, reflection, and absorption based on material properties and environmental conditions. This enables the reproduction of realistic sensor behavior in scenarios such as parking maneuvers, obstacle detection, or close-range navigation. By incorporating accurate intrinsic and extrinsic parameters, the simulation can replicate manufacturer-specific sensor characteristics, ensuring that virtual signals closely match those of real hardware.

Over-the-air sensor stimulation can be seen as the interface between the simulation and real sensors that are mounted on the vehicle. It automatically analyzes the ultrasound signal generated by the vehicle's internal ultrasonic sensor, including ultrasonic frequency and signal amplitude. Based on the simulated environment, the response is calculated in AURELION and precisely reproduced and injected as a delayed echo signal to stimulate the sensor installed on the vehicle. By doing so, the over-the-air sensor stimulation delivers the appropriate feedback signal back to the real sensor, ensuring that the sensor perceives the simulated environment as if it were happening in the real world.
This closed-loop setup enables deterministic testing of ultrasonic sensors under a wide variety of conditions. It allows engineers to validate sensor behavior in complex scenarios, such as parking maneuvers, obstacle detection, or close-range navigation, while maintaining full reproducibility and scalability.
 

Summary

Vehicle-in-the-loop (VIL) testing enhances the validation of automated parking assistant systems by combining real vehicle hardware with a simulated environment. Using AURELION and ASM models, realistic ultrasonic responses are calculated and transmitted via over-the-air (OTA) stimulation back to real sensors, enabling closed-loop testing. 

VIL testing is key for validating automated parking assistants because it combines real vehicle hardware with a virtual environment. This enables realistic testing under controlled conditions without the high costs and risks of physical road tests. VIL offers high flexibility for complex scenarios, supports edge case and safety evaluations, and accelerates software update cycles. In the context of future virtual homologation, VIL is particularly relevant as it provides reproducible evidence of safety and functional compliance, paving the way for digital type approval processes.