Innovation project for safeguarding autonomous vehicles

The Bavarian Ministry of Economic Affairs, Regional Development and Energy (StMWi) is funding the new technology project "Holmes3" of the Software Engineering Laboratory for Safe and Secure Systems (LaS³) of the East Bavarian Technical University Regensburg (OTH), Elektronische Fahrwerksysteme GmbH (EFS) and imbus AG. The aim of the project, which has just been officially launched, is to further develop methods for testing and securing autonomous vehicles - one of the greatest challenges for the functional safety of autonomous vehicles.

With the initiative "Artificial Intelligence - Autonomous Mobility", the Bavarian Ministry of Economic Affairs, Regional Development and Energy (StMWi) promotes innovations in the fields of artificial intelligence (AI) and data science in application areas such as mobility or logistics. The Software Engineering Laboratory for Safe and Secure Systems at OTH, Elektronische Fahrwerksysteme GmbH (EFS) and imbus AG have joined forces to form a team for this technology project.

As part of the project, a manufacturer-independent approach and tool environment for scenario-based testing and for AI-based analysis of driving scenarios using a "Causal Inference Engine" will be developed and made available for use.

The technology project will run until mid-2023, and the solution resulting from this Bavarian initiative will subsequently be made available to companies from other industries that develop or operate intelligent, autonomous systems.

"When testing classic software or IT systems, the test environment is usually constant or limited to a few predefined variants. Testing safety-critical AI-based autonomous systems presents software testing with the challenge of running through and systematically testing as many as possible of the countless traffic scenarios that come into question or that are conceivable" explains Tilo Linz, CEO of imbus AG. "Scenario-based testing will define the basis for an international standard here in the future as a new test method in conjunction with the automotive industry's "ASAM Open-SCENARIO"."

"Our claim is to technologically advance the development of autonomous systems. We can use causal discovery methods to describe how autonomous vehicles behave in critical situations," said Daniel Ebenhöch, responsible product owner at EFS. "Modeling causal relationships in a comprehensible way - we bring this expertise significantly to the project. I personally am very much looking forward to the cooperation."

Prof. Dr. Jürgen Mottok quotes Wilhelm Busch and adds, "Those who watch see more than those who play along. Thus, similar to the specially empowered character Sherlock Holmes imagined by Sir Arthur Conan Doyle, the R&D project HolmeS³ will develop the proven ability of "seeing" with causal inference. The art, then, is to correctly interpret behavior causally. This enables classification in the Safety Integrity Level (SIL) of functional safety."

The project, which is funded by the StMWi, will sustainably strengthen Bavaria as a business location and the competitiveness of the Bavarian automotive industry in this highly innovative and forward-looking field.

More information on the HolmeS³ technology project is available at:

https://www.imbus.de/en/forschung/holmes3
https://www.las3.de