Autonomous driving can work well in precisely defined areas of use. Yet if driverless cars are let loose into the free-for-all of everyday road traffic unprepared, difficulties can crop up that these vehicles are not prepared to cope with. Three examples.

Complex behavior presented by modern vehicles relies on complex sensors recognizing the environment of the vehicle. Unfortunately, this functionality does not operate in a black-white scheme.

High-performance computers (HPC) as the central elements of E/E-architectures in software-driven vehicles need an adequate operating system solution. Linux is a good choice for HPCs.

The demanding integrity and availability requirements of Highly Automated Driving need to be addressed by suitable architectures.

The days of hollow “women's power” phrases are over. This has long since become a social demand for equal opportunities for women, which is clearly formulated, emphatically demanded and often put into practice - albeit with varying degrees of commitment. Jessica Kelly, who has been a researcher at the institute since 2022, explains what it looks like in science and at Fraunhofer IKS.

Algorithms used in automated driving systems are complex and use non-deterministic deep neural networks. Some variants of deep neural networks can be explained theo­retically but their behavior under all conditions for the set of safety-critical scenarios to be covered by the automated driving system in the given operational design domain is generally not readily understood. Here is how it could work – a new approach.

Research teams at the Fraunhofer IKS live and breathe diversity. The researchers come from a total of 25 countries. Dr. Alexandre Sawczuk da Silva from Brazil is one of them. In this interview, he describes his path to Fraunhofer research.

Artificial intelligence and LLMs in particular are seen by many as a beacon of hope for the overburdened healthcare system. Above all, AI-based automation could quickly provide relief for knowledge management routine tasks. Until that happens, problems with security and safety must be solved and legal requirements fulfilled. Fraunhofer IKS research is addressing both of these issues.

Industrial production is undergoing radical change. With constant technological advancement and societal change, industrial automation must also reinvent itself to meet changing requirements and challenges – with the keyword here being adaptive production.

The performance of machine learning (ML) models is typically assessed using metrics such as accuracy, precision, recall, F1 score, etc. But how trustworthy are these single point estimates really? Such metrics reflect reality only partially and can provide us with a false sense of safety. Especially in safety-critical areas, this can be dangerous. Fraunhofer IKS develops a formal framework for the quantification of uncertainty which aims to make the residual risk “more computable”.