Research Project AD@chronVAL
Research Project AD@chronVAL
Ensuring that today’s increasingly complex automotive software is functionally safe is more complicated than ever. As well as showing that such systems function as expected, they must also demonstrate that they can respond within predefined time limits under all possible conditions. Pushing this requirement are the latest generation of autonomous driving (AD) functions and advanced driver-assistance systems (ADAS) that intervene in the process of driving. Mathematical analysis will be advantageous in analyzing such complicated systems as it can find the worst-case conditions that highlight the corner cases of the system’s execution.
These future AD and ADAS capabilities will place new requirements on worst-case analysis tools. By bringing together their recent scientific research in this area, INCHRON and Ulm University will leverage their mathematical real-time analysis capabilities resulting in improvements for chronVAL, part of INCHRON’s chronSUITE, in the AD@chronVAL research project.
The goal is to provide industry users with a real-time analysis tool that delivers sufficiently accurate, comprehensive, reliable, and transparent results for the multitude of practically relevant issues arising in future autonomous driving applications. From the user’s point of view, these should be on a par with the corresponding simulation results.
A key focus of the research will be on the analysis of event chains, which are essential to the analysis of end-to-end data flows in the architectural design and testing of AD and ADAS systems. The research project will be carried out in close cooperation with lead customers by allowing them to contribute their planned use cases. As a result, they will directly benefit from the research results.
Additional research priorities include the development of meaningful visualizations of the results and calculations, together with close integration of the mathematical analysis with the simulation capabilities of chronSIM. This will enable users to benefit from the combined strengths of the respective methods. Against the background of today’s development approaches, this will help industrial developers who, from experience, do not have the time to deal with new mathematical methods.
The AD@chronVAL project will deliver a sophisticated mathematical analysis process and lead to improvements in visualization and the intelligibility of analyses within chronVAL. Undertaken in close cooperation with Ulm University, the project benefits from INCHRON’s more than 18 years of experience in timing in customer applications coupled with the theoretical knowledge of the Institute for Embedded Systems/Real-Time Systems’ team. The project is supported financially by KMU-innovativ, a program of the the German Federal Ministry of Education and Research (BMBF, Bundesministerium für Bildung und Forschung) for SMEs.
April 2022 Project Update
After sharing the goals of the AD@chronVAL project, the team has been inundated with offers of use cases and other insights from automotive industry experts and teams. This is currently being evaluated to understand how it can be applied to the project.
Such feedback is invaluable for the development of new timing analysis methods to ensure future requirements will be supported.
Today, the team is working intensively with a range of industry partners to understand how they can best benefit from the new methodologies and tools being explored. Due to the phase of the project, there is still time to accept and integrate further use cases and industry challenges around timing analysis.
If you have something you would like to submit for consideration, please get in touch with us via your usual INCHRON contact or use the form below.
The AD@chronVAL project depends on input from real-world projects. If you are currently engaged in autonomous driving solutions or ADAS applications and are willing to participate, we’d be pleased to hear from you. Please fill out the form below, and our research team will be in touch shortly with more details on how to contribute.