System Health Monitoring for Autonomous Systems

Working together for intelligent mobility

With the growing complexity of software in vehicles, the risk that different solutions from different providers will not be compatible with each other is increasing. For this reason, a worldwide development partnership of vehicle manufacturers, suppliers, service providers and companies from the automotive electronics, semiconductor and software industry was established with AUTOSAR. The primary goal of this partnership is to jointly develop and establish an open industry standard for automotive E/E software architecture.

By standardizing common methodology, basic system functions and functional interfaces, development partners can integrate, exchange, re-use and transfer functions within a vehicle network and substantially improve their efficiency of development.

In addition to Fraunhofer IKS, Fraunhofer FOKUS and Fraunhofer AISEC are also involved as members.

Exploration of standardized solutions

As part of this collaboration, Fraunhofer IKS together with other members conducts research mainly in the practical application of System Health Management. In addition, the scientists of Fraunhofer IKS evaluate concepts and research new solutions on adaptive and autonomous systems that comply with the standard and can thus be directly transferred to industrial practice.

Currently, Fraunhofer IKS is part of two working groups:

Working Group Safety (WG-SAF)

One of the main challenges with autonomous systems is an appropriate diagnosis and safety management to ensure system health and operability. To solve these challenges, systems need to be aware of malfunctions and potential issues of single sensors or modules within the systems. The WG-SAF pursues the goal of improving and standardizing the System Health Monitoring (SHM) across various platforms. By evaluating health information from multiple platforms, health indicators are created that describe the health of a subsystem. For instance, this makes it possible to react optimally to the failure of vehicle functionalities.

Working Group Application Interfaces (WG-AIF)

The WG-AIF is working on the standardization of automated driving sensor interfaces. In the future, Fraunhofer IKS sees potential in a higher degree of automation in design and validation and verification activities. This is to be achieved with a strong linkage between model-based safety analysis and assurance cases with actual implementation.

 

Research by Fraunhofer IKS

© iStock.com/Just_Super

Fraunhofer IKS has also been involved in AUTOSAR projects together with various companies on several occasions in the past:

  • Support for AUTOSAR Log and Trace development and standardization
  • Implementation of AUTOSAR BSW SoAd
  • AUTOSAR Ethernet/TCP/IP Stack Conformance Testing

As part of this cooperation, Fraunhofer IKS researchers are able to apply experience gained from previous projects in the field of autonomous driving and Industry 4.0.

 

Fault-tolerant Embedded Platforms

Embedded systems cannot be allowed to simply shut down when outages occur. The research of Fraunhofer IKS ensures the functional safety of these platforms by developing architectures that provide greater flexibility for safety-critical applications.

 

Adaptive Safety and Performance Management

Conventional validation approaches are unsuitable for dynamic contexts. Therefore, Fraunhofer IKS develops solutions designed for safety-critical systems. In contrast to conventional methods, this approach specifically addresses the three interconnected issues of safety, reliability and cost.

 

What happens when AI fails to deliver?

Artificial intelligence (AI) has to be able to handle uncertainty before we can trust it to deliver in safety-critical use cases, for example, autonomous cars. The Fraunhofer IKS is investigating ways to help AI reason with uncertainty, one being the operational design domain. Read more on our Safe Intelligent blog.