Software company Elektrobit has announced it is working with Infineon Technologies, a leading semiconductor supplier for autonomous driving, and NVIDIA, a specialist in visual computing, to make it easier for the automakers and their suppliers to create advanced driver assistance systems.
Elektrobit said that the companies are delivering a “first-of-its-kind solution that represents a powerful development platform for automated driving”.
It is claimed the offering integrates technologies from EB, Infineon and NVIDIA to deliver the increased computing power and intelligence required to handle the complexity of automated driving. Moreover, it is claimed, ADAS applications developed using the solution will enable a vehicle to not just sense, but to interpret what is happening around it and communicate information to the driver, to other critical systems within the vehicle, to other vehicles nearby, and to the cloud.
The solution consists of the NVIDIA DRIVE PX self-driving computer integrated with EB's AUTOSAR 4.x-compliant EB tresos software suite, which runs on the NVIDIA Tegra processor and Aurix™ 32-bit Tricore™ microcontroller from Infineon. The AURIX microcontroller features high real-time processing performance and enhanced embedded safety and security features that make it well-suited for use in ADAS applications. The NVIDIA DRIVE PX platform enables the development of systems that capture and process multiple HD camera and sensor inputs and provides a rich environment for advanced graphics, computer vision and machine learning. EB tresos software provides seamless integration capability of Linux and AUTOSAR applications as well as NVIDIA DRIVE™ PX -specific basic functionality for monitoring and redundancy management. It enables cross-CPU-communication with a safe and reliable execution environment, supporting the highest automotive safety integrity level (ASIL).
"For the first time, EB has developed an architecture software package based on our AUTOSAR network communication that will enable automakers and their suppliers to define the future of mobility," said Björn Giesler, senior director driver assistance, EB. "Automated Valet Parking is a typical use case for this platform. Such an automated driving feature processes a vast amount of data and requires a lot of computing power for a high level of safety. Our integrated hardware-software solution provides the easiest and most cost-effective way for automakers to build sophisticated ADAS features," said Giesler.
"Autonomous driving requires a massive amount of energy-efficient computing resources, plus the highest safety standards and software reliability," said Sahin Kirtavit, senior director of automotive ecosystem, NVIDIA. "With this solution, we are bringing the power of supercomputing into the vehicle."
"Infineon provides key technologies on the road to achieve the ultimate goal of fully autonomous driving," said Hans Adlkofer, vice president automotive systems, Infineon Technologies AG. "The microcontroller AURIX is the trust anchor for the entire board net architecture due to its high real-time processing performance combined with its embedded safety and security features and highest flexibility to serve the different advanced driver assistance systems in the autonomous car. As a result, AURIX and powerful safety software solutions such as the EB tresos suite form the safe and secure hub for taking safety critical autonomous driving strategy decisions and communicate with the actuator systems of the car."