—
At the forefront of innovation in autonomous vehicle technology, NVIDIA has made a significant announcement regarding its autonomous vehicle platform, NVIDIA DRIVE AGX Hyperion. This platform, which is designed to propel the future of self-driving vehicles, has achieved crucial safety certifications from TÜV SÜD and TÜV Rheinland, two of the most respected authorities in automotive safety and cybersecurity. This milestone sets a new standard for safety and performance in the realm of autonomous vehicles.
A Comprehensive Autonomous Driving Platform
NVIDIA DRIVE Hyperion stands out as the industry’s first and only end-to-end solution for autonomous driving. This platform includes several key components: the DRIVE AGX system-on-a-chip (SoC) and a reference board design, the NVIDIA DriveOS automotive operating system, an advanced sensor suite, and an active safety and level 2+ driving stack. These components work together seamlessly to provide a robust foundation for developing self-driving vehicles.
Major automotive pioneers, such as Mercedes-Benz, Jaguar Land Rover (JLR), and Volvo Cars, are adopting this platform due to its modular design. This allows manufacturers to select and utilize only the components they require, ensuring flexibility and efficiency in their development processes. Moreover, the platform is scalable and engineered to be compatible with future generations of DRIVE SoCs, ensuring longevity and adaptability in an ever-evolving technological landscape.
The Next Phase: High-Performance NVIDIA DRIVE AGX Thor
The latest iteration of the DRIVE Hyperion platform, expected to be available in the first half of this year, is designed for both passenger and commercial vehicles. It will feature the DRIVE AGX Thor SoC, which is built on the NVIDIA Blackwell architecture. This architectural advancement promises to deliver exceptional performance and capabilities, setting the stage for a significant leap forward in autonomous vehicle technology.
Jensen Huang, the founder and CEO of NVIDIA, highlighted the transformative impact of this platform: "A billion vehicles driving trillions of miles each year move the world. With autonomous vehicles — one of the largest robotics markets — now here, the NVIDIA Blackwell-powered platform will shift this revolution into high gear." He further emphasized the role of physical AI world foundation models in understanding and interacting with the real world, stating that NVIDIA DRIVE is purpose-built for this new era, offering unmatched functional safety and AI capabilities.
Certified Safety and Cybersecurity for Next-Generation Vehicles
As vehicles become increasingly software-defined, with the ability to receive new features and functionalities over their lifetimes, ensuring their safety and cybersecurity becomes paramount. NVIDIA has leveraged its 15,000 engineering years invested in vehicle safety to develop DRIVE Hyperion, ensuring that advanced automotive systems with rich, AI-based functionalities comply with the industry’s stringent safety and cybersecurity standards.
NVIDIA has recently received several important safety certifications and assessments from accredited third parties. TÜV SÜD has granted NVIDIA the ISO 21434 Cybersecurity Process certification, which covers its automotive SoC, platform, and software engineering processes. Additionally, NVIDIA DriveOS 6.0 conforms to ISO 26262 Automotive Safety Integrity Level (ASIL) D standards, with certification pending release.
TÜV Rheinland conducted an independent safety assessment for the United Nations Economic Commission for Europe (UNECE), focusing on NVIDIA DRIVE AV and its compliance with safety requirements for complex electronic systems. These certifications underscore NVIDIA’s commitment to maintaining the highest standards of safety and cybersecurity in its automotive technologies.
Furthermore, NVIDIA has been accredited by the ANSI National Accreditation Board (ANAB) to provide safety and cybersecurity inspections for its DRIVE ecosystem partners. The newly established NVIDIA DRIVE AI Systems Inspection Lab will assist the DRIVE automotive ecosystem in building autonomous driving software that meets evolving safety and AI standards.
Pioneering Independent Accreditation in Automotive Safety
NVIDIA is the first platform company to receive a comprehensive set of third-party assessments for its automotive technologies, which include the NVIDIA DRIVE end-to-end self-driving platform. This platform spans SoC, OS, sensor architecture, and level 2+ application software. Moreover, NVIDIA has achieved independent accreditation as an AI systems safety and cybersecurity inspection lab for the automotive market, further solidifying its leadership in this field.
Advanced Computing Power with NVIDIA DRIVE Thor
At the heart of the DRIVE Hyperion platform is NVIDIA DRIVE Thor, the core computer that succeeds the production-proven NVIDIA DRIVE Orin. The compatibility and scalability of its architecture allow developers to leverage existing software from earlier DRIVE product generations and integrate future updates seamlessly, facilitating smooth development pipelines.
DRIVE Thor is based on the NVIDIA Blackwell architecture and is optimized for handling the most demanding processing workloads. These include tasks involving generative AI, vision language models, and large language models. Its architecture is designed to enhance generalization, reduce latency, and boost safety, harnessing powerful NVIDIA accelerated computing to run both the end-to-end AV stack and a proven safety stack in parallel.
This innovation paves the way for the next era of autonomous vehicle technology, known as AV 2.0, which aims to deliver humanlike autonomous driving capabilities for navigating complex roadway scenarios.
Supporting Development with Three Key NVIDIA Systems
In addition to the DRIVE AGX in-vehicle computer, NVIDIA offers two other essential computing systems for automotive-grade AV development. The NVIDIA DGX systems are used for training advanced AI models and building a robust AV software stack in the cloud, while the NVIDIA Omniverse platform, running on NVIDIA OVX systems, is employed for simulation and validation. These three systems, now enhanced with the NVIDIA Cosmos world foundation model platform, are poised to accelerate end-to-end AV development and facilitate mass deployment.
Conclusion and Further Exploration
NVIDIA’s comprehensive approach to automotive development, underpinned by its three-computer strategy and the Cosmos world foundation model platform, represents a significant advancement in the field of autonomous vehicles. For those interested in learning more about these developments and other automotive news, Jensen Huang’s CES opening keynote provides valuable insights. You can access it through NVIDIA’s official event page.
In conclusion, NVIDIA’s latest advancements in the autonomous vehicle sector highlight the company’s commitment to safety, innovation, and performance. As the industry continues to evolve, NVIDIA’s DRIVE Hyperion platform and its associated technologies are set to play a pivotal role in shaping the future of transportation.
For more detailed information on NVIDIA’s latest developments and to stay updated on their ongoing advancements, visit their official blog.
For more Information, Refer to this article.
