How can autonomous cars, lorries, buses and trains drive safely and reliably when the traffic situation is constantly changing due to unexpected obstacles or changing weather conditions? This question is at the centre of the CONTROL research project, which is funded by the Federal Ministry for Economic Viability and Energy (BMWE) for three years with a total of 15.6 million euros. The consortium involves 24 partners from industry and the scientific community. Ulm University is also involved with a funding contribution of more than half a million euros. Coordinated by CONTROL is coordinated by Siemens AG and Valeo Schalter und Sensoren GmbH.
Autonomous driving is regarded as a key technology for the mobility of the future. It promises more comfort, greater fleet efficiency, lower emissions, a more sparing use of resources and, above all, new mobility options - for example through autonomous vehicles for passenger transport as well as driverless trains and lorries. The automated systems that are already in use today only work under clearly defined environmental conditions. What is missing is the ability to operate safely in the open, highly dynamic world with changing external conditions. This is precisely where the CONTROL project comes in. The acronym stands for "Controlling Risk of Highly Automated Transportation Systems Operating in Complex Open Environments".
"Our joint research addresses difficult scenarios that can also be challenging for autonomous vehicles on the road or rail. These include unexpected obstacles on the road or track as well as poor visibility, for example in changeable weather or problematic weather and lighting conditions," explain consortium leaders Dr Cornel Klein (Siemens) and Dr Sanwardhini Pantawane (Valeo). In this context, experts speak of so-called long-tail events. This refers to rare critical events or scenarios that are difficult or impossible to test using data sets or test drives.
Over the next three years, the research and development teams at CONTROL will develop innovative methods for safeguarding autonomous vehicles that will help to safely manage such long-tail events - for cars, lorries and trains. The development of a safety argumentation that systematically records, evaluates and controls uncertainties in vehicle operation and integrates them into the architecture of autonomous systems is central. To this end, metrics, models and tools are being developed that are scalable, cross-domain and industrially realisable. The entire perception chain, from sensor technology to data processing and evaluation, is considered as a coherent system. Influences such as weather, environment or technical faults are modelled and taken into account using adaptive sensor-based processes. The reliability and functionality of the safety architecture is tested and evaluated in computer simulations, in laboratory set-ups such as test stands and in demonstrators and test vehicles.
Germany is a pioneer in safety technologies for automated driving
Highly automated vehicles work with a large amount of data from a wide variety of measuring systems such as sensors or cameras. This data must be collated and processed centrally so that the vehicle's central computing and control unit can derive sensible driving manoeuvres on this basis. Scientists at Ulm University are researching quality metrics for data fusion in CONTROL and are receiving 570,000 euros from the funding pot for this. "We are developing methods that enable the system itself to assess the quality of the data it generates and thus provide information as to whether this is sufficient for it to be able to act safely," says Professor Michael Buchholz from the Institute of Measurement, Control and Microtechnology, who is involved in the CONTROL project with a working group.
"The results are immediately transferred from research to practice. The consortium project reinforces the competitiveness of German industry in the field of autonomous mobility and creates an important foundation for new products and standards - from road and rail to other fields of application such as industrial automation or robotics," the project managers are convinced. This will ensure that Germany retains its pioneering role in automated driving technology in the future.
About CONTROL
The research project is being funded by the Federal Ministry for Economic Viability and Energy (BMWE) with a total of 15.6 million euros. CONTROL brings together 24 partners from industry and science - including vehicle manufacturers, suppliers for the road and rail sectors, universities and research institutions. Siemens AG and Valeo Schalter und Sensoren GmbH are leading the consortium.
The project is supported by the German Association of the Automotive Industry via the "VDA Leitinitiative autonomous and connected driving". The flagship initiative promotes so-called pre-competitive consortium projects of German car manufacturers and suppliers who - although they compete for market share - conduct research on key technologies and cross-cutting topics such as AI, safety and standardisation together with players from the scientific community.
Further information can be found at www.control-projekt.de
Contact consortium leader:
Dr Cornel Klein (Siemens), e-mail: cornel.klein@siemens.com
Dr Sanwardhini Pantawane (Valeo), e-mail: sanwardhini.pantawane(at)valeo.com
Contact University of Ulm:
Prof. Dr Michael Buchholz, e-mail: michael.buchholz(at)uni-ulm.de
Text: CONTROL Consortium + University of Ulm
A list of the research and industrial partners of the CONTROL consortium project can be found HERE!!!