This page was machine-translated and may differ from the original. View original

NVIDIA Unveils Software Platform for Autonomous Vehicles

Google 우선 소스Published2019.12.19 11:10
Drive AGX Orin, 7x improvement in SoC compared to previous generation
Scalable from Level 2 to Level 5, compatible with architecture


NVIDIA has unveiled NVIDIA DRIVE AGX Orin, a cutting-edge software-defined platform for autonomous vehicles and robots.

▲ NVIDIA DRIVE AGX Orin <Photo = NVIDIA>

This platform was launched based on a new system-on-chip (SoC) called Orin, which consists of 17 billion transistors.

The Orin SoC integrates NVIDIA's next-generation GPU architecture, Arm's Hercules CPU cores, and a new deep learning and computer vision accelerator that provides 200 trillion operations per second, delivering 7x performance improvement over the previous generation NVIDIA Xavier SoC.

Orin is designed to handle numerous applications and deep neural networks operating simultaneously in autonomous vehicles and robots while meeting safety standards such as ISO 26262 ASIL D.

Built as a software-defined architecture, NVIDIA Drive AGX Orin is an architecture-compatible platform scalable from Level 2 autonomous driving to Level 5 fully autonomous driving vehicles, enabling OEMs to develop large and complex software suites.

Orin and Xavier both have open CUDA, TensorRT APIs and librariesBecause it is programmable, developers can utilize it efficiently across multiple product generations.

NVIDIA Founder and CEO Jensen Huang stated, “The biggest computing challenge facing our society today is building safe, autonomous vehicles,” adding, “As the investment required to implement autonomous vehicles increases exponentially and the complexity of the task grows, the need for scalable and programmable software-defined AI platforms like Orin will become even greater.”

The NVIDIA Drive AGX Orin product family includes various configurations based on a single architecture and is aligned with automotive manufacturers' 2022 production schedules.
본 기사에 대한 정정·반론·추후보도 청구는 보도 청구 안내를, 그간 게재된 보도문은 정정·반론보도 모아보기를 참고해 주세요.
최인영 기자