Testing What Really Matters

IAV Valdivia Sample

Safety validation for automated vehicles can be quite challenging. We have developed IAV Valdivia Sample to test Autonomous Driving (AD) functions more efficiently and reliably using machine learning (ML) methods. It helps you to comprehensibly identify the truly relevant scenarios out of millions of possible test scenarios.


Save up to 80%

on testing effort

Automated vehicles must safely manage countless scenarios. Given the multitude of potential situations, it is not possible to test all of them. This is where IAV Valdivia Sample comes in. The tool assists in the targeted selection of test scenarios and helps to reliably test automated driving functions in the field of ADAS and AD. With this approach, customers can save up to 80% on testing effort*.

*exemplary calculation based on research results




What is IAV Valdivia Sample?

How do you benefit from IAV Valdivia Sample?

Generate traceable results

IAV Valdivia Sample uses a scientifically grounded approach with state-of-the-art probabilistic ML models. In this way, IAV Valdivia Sample offers a transparent process and traceable decisions.

Save time and reduce costs

The software provides the opportunity to reduce your test effort by up to 80% through a lower number of simulations and less computation time and licenses required. This leads to increased efficiency and cost savings.

Whether in the cloud or locally - just get started!

IAV Valdivia Sample is designed so that it can be integrated into fully automated test environments (SiL or HiL) via REST APIs. The solution is cloud-ready and can be used as software as a service.

Countless use cases for numerous industries

Originally designed for use in the AD sector, the tool can be used wherever efficient model building methods are required for the development or approval of complex systems - e.g. in the energy industry or aviation.


How does it work?

With IAV Valdivia Sample, IAV has developed a software that uses statistical techniques and artificial intelligence to generate test scenarios. This process is iterative and adapts independently to the previously unknown system and its properties – because every system has its own weaknesses that need to be uncovered.


Generate different scenarios

The test scenarios are initially generated using statistical techniques to achieve a broad coverage and consider very different scenarios.


AI comes into play

The results of these tests are then analyzed using AI to determine which properties of the scenarios, alone or in combination, favor critical outcomes. This could be, for example, the interaction of a low sun and a wet road.


Iterations lead to a better understanding

Finally, critical scenarios for the system are generated iteratively and their data is analyzed again. With each iteration, more weaknesses can be uncovered and the system can be better understood.

Using artificial intelligence, this procedure goes beyond traditional optimization and experimental design methods. It is characterized by the use of probabilistic metamodelling and repeated, adaptive trial planning (Adaptive Importance Sampling). The probabilistic models can ultimately be used to calculate probabilities of critical scenarios, providing a measurable basis for the release decision of safety-critical systems.



Everything at a glance

with the IAV Valdivia Sample Dashboard

Better together

Our partnerships


Let’s meet

13 May 2024, 10 AM Mobex-Webinar Online Link
14 – 16 May 2024 Safety Week Hanau, Germany Link
22 – 24 May 2024 JSAE Conference Yokohama, Japan Link
04 – 06 June 2024 Automotive Testing Expo Stuttgart, Germany Link




Are you curious now?

Ask Mike

Mike Hartrumpf
Technical Consultant Functions & Simulation