Simulation is bound to play a key role in homologation of automated vehicles in the future, given the countless potential driving scenarios that need to be tested. To validate simulation as a test method in vehicle approval, TÜV SÜD, US AI computing company NVIDIA Corporation, and Austria-based drive system developer AVL GmbH are now embarking on a far-reaching collaboration. Their objective is to define safety requirements, critical driving scenarios, and the necessary evaluation criteria in order to establish simulation as a testing tool. With this new collaboration, TÜV SÜD underscores its claim to be the leading independent third-party testing organization in the development of autonomous driving.
“Using technical means to ensure that automated driving functions are executed accurately, reliably, and safely is one of the greatest challenges in the homologation and approval of autonomous vehicles,” says Patrick Fruth, Head of the Mobility Division at TÜV SÜD. “TÜV SÜD has always been a guarantor of safety – and will remain so in the future. Together with NVIDIA and AVL, we look forward to making tomorrow's mobility safe and creating the basis for approval.”
Validating new test methods
In addition to conventional driving tests, simulation will play a central role in the future. After all vehicles must react correctly, reliably, and safely in every possible driving situation, and TÜV SÜD experts estimate that each fully automated driving function comprises 100 million such situations. Given this, scalable verification and test methods are needed to cope with this enormous number of scenarios. Simulation of driving situations is a method that makes these goals attainable. TÜV SÜD, NVIDIA, and AVL are working together to validate and establish simulation as an approval tool. “Together, we are tackling the issue of what digital homologation may look like in the future and how it can be implemented,” says Dr. Houssem Abdellatif, Global Head Autonomous Driving at TÜV SÜD. The plan is to apply the findings as early as the vehicle development phase in the future. “Such early consideration of new discoveries not only improves efficiency, but also ensures that driving functions will demonstrably comply with regulatory requirements right from the outset,” says Dr. Abdellatif.
Connecting the virtual and real worlds
Tasks are distributed among the collaboration partners. AVL builds the bridge to testing real vehicles by running the physical vehicle on a test bench for autonomous driving called AVL DRIVINGCUBE™.
Georg List, VP of Corporate Strategy at AVL, said: “In this cooperation, we connect simulation all the way through to vehicle-in-the-loop to provide an efficient development and validation environment. One benefit of such a solution is to expose vehicle electronics, installed in a real car, to virtual traffic situations from our scenario generator via our co-simulation solution Model.CONNECT. This hybrid virtual-real validation enables us to test a car here for rush hour conditions on Fifth Avenue in New York, and thus develop a digital twin with high correlation to real driving tests.”
TÜV SÜD's main task involves defining safety requirements, examples of critical driving scenarios, and the necessary evaluation criteria. In addition, TÜV SÜD is responsible for evaluating the trustworthiness of the test equipment used, including the simulation tools and the test bench. “Trustworthiness of the equipment must be ensured to enable us to use the test results in type approval,” explains Christian Gnandt, Head of Virtual Methods at TÜV SÜD.
The virtual world comes from NVIDIA. The graphics and AI computing company from Silicon Valley supplies the automated driving hardware and software and makes its DRIVE Constellation platform and DRIVE Sim software available. Serkan Arslan, Director of Automotive at NVIDIA, says: “The ability to drive millions, if not billions, of miles in VR is critical to bringing safe, automated driving solutions to market. With TÜV SÜD, we have a renowned partner with an international reputation for vehicle and traffic safety.”
Ensuring safety right from the start
TÜV SÜD has accompanied the development of highly automated driving as a safety partner in a wide variety of areas right from the start. In the PEGASUS project of the Federal Ministry for Economic Affairs and Energy, for example, the experts work with 16 partners from industry and science to establish the requirements for methods and tools used in safeguarding highly automated driving functions. In the "Driver Assistance Systems" special committee at the Federal Ministry of Transport, TÜV SÜD's experts are assisting in the revision process for the relevant standards.
Main areas of focus in connection with autonomous driving include functional safety, cybersecurity, and the handling of sensitive data in general. TÜV SÜD likewise has expertise in these fields, and is driving research there by partnering industry enterprises in a wide variety of projects. One such example is the development of a “TÜV” or testing method, for algorithms, in collaboration with the German Research Center for Artificial Intelligence (DFKI).
As another example of TÜV SÜD's commitment to the mobility of tomorrow, the company partners the test field operator for autonomous driving in the German state of Baden-Wuerttemberg and KVV, Karlsruhe’s public transportation authority, to operate a test field where research institutions and companies have the opportunity to test their developments in everyday road traffic.
Patrick Fruth: “Where approval for road traffic is concerned, the octagon is always on board – an important aspect in establishing standards for the approval of highly automated vehicles.”
Press-contact: Vincenzo Lucà