The ZF Group, a specialist in mobility systems, has collaborated with Cognata, a company specializing in autonomous systems and artificial intelligence. On July 1, the two companies announced the launch of ADAS.ai, a set of data and AI-based services for the validation of driver assistance systems (ADAS). With this new tool, ZF wants to help manufacturers accelerate the development phase of ASAS for trucks and passenger vehicles. The solution can be used for ADAS systems developed by ZF or other Tier 1 suppliers.
ADAS.ai
platform to accelerate the development phase of driver assistance systems
With the help of Cognata, ZF has developed a scalable set of AI and data-based services to digitally validate driver assistance systems. NamedADAS.ai, the platform is based on two major innovations:
- A set of very high resolution data, synchronized from multi-sensors. This data set is representative of all the scenarios and mileages required to validate ADAS systems of Level 2 or higher on a global scale.
- AI technology (developed in collaboration with Cognata) that transforms all of the very high resolution data into input data for new sensors being considered for new vehicle applications.
Christophe Marnat, Executive Vice President of ZF’s Electronics and ADAS division, discusses the use of data in his firm’s proposed solution set:
“We use real data in an end-to-end approach, because we know the needs of our customers and regulators. With our approach, a manufacturer needs neither a large number of test pilots nor large budgets for confident ADAS development and validation.”
The ZF Group vice president also elaborates on how the new service portfolio works and what it is for:
“We bring together the real and virtual worlds by using real data to manage both modalities. Our goal is to enable full virtual engineering of ADAS systems for all Tier 1 OEMs. The OEMs want the ADAS features to be fully validated digitally in the same way that crash test validations are conducted, and therefore before the first prototype is built. And ZF has the technology to make this happen, with the result that digital validation times and costs can be reduced by up to 20 percent.”
A set of secure services designed with artificial intelligence
ADAS.ai was custom designed and incorporates proprietary algorithms from Cognata. Danny Atsmon, CEO of the company, adds:
“The AI developed by Cognata enables the ZF ADAS.ai solution to transfer real-world driving data to all existing and new vehicles with ADAS technology up to Level 2+ and validate it in the virtual: it’s the transformation from real to real.”
To design the platform, ZF has leveraged all the knowledge it has in the world of ADAS hardware and software technologies, whether in the validation processes or in the development of documentation for compliance related to existing regulations around the world. Security remains an important aspect in the eyes of Christophe Marnat:
“Our priority is the safety and reliability of today’s and tomorrow’s ADAS technology. This is reflected in our development and validation process, which is not based on simulations but on real driving data collected over hundreds of thousands of kilometers worldwide. Indeed, ZF knows what it takes to make an ADAS system operational.”
According to ZF Group, ADAS.ai will offer significant cost and quality advantages over traditional engineering and validation processes based on existing physical testing and simulation technologies.
Translated from ZF et Cognata s’associent pour accélérer la validation des systèmes d’aide à la conduite grâce à l’IA
