Nvidia CEO Jensen Huang said the company’s latest architecture will enable a vehicle to be upgraded to Level 5 autonomy.
Nvidia is expanding the use of its chip technology to allow for the use of software to alter a vehicle’s advanced driver assistance systems from simply help for drivers to Level 5 autonomy with over-the-air updates.
The company, which made its name making hardware used by the world’s most intense gamers, has become a force in the automobile industry by transferring its graphics and know-how into cockpits and center stack of cars and trucks and is moving it closer to automated driving.
Jensen Huang, Nvidia founder and CEO, recently announced the expansion of the Drive AGX platform, using new variants of the upcoming Orin system-on-a-chip and new Nvidia Ampere GPUs. With a single architecture, manufacturers can deploy a high-performance AI system to make every vehicle in their line-up software-defined, Huang said.
(AV testing miles in California jumped 40% in 2019.)
Karma’s teamed with Nvidia to create a SAE Level 4 autonomous concept van, called the L4 Autonomy.
Nvidia’s Ampere-based platform is now capable of handling higher resolution sensor inputs and more advanced autonomous driving inputs required for full self-driving robotaxi operation.
“The architecture offers the largest leap in performance within the eight generations of Nvidia GPUs — boosting performance by up to six times,” Nvidia officials said
Tests will begin next year and the system available for automakers starting production in late 2022, laying the foundation for the next-generation of the programmable, software-defined Nvidia Drive AGX line-up.
(Americans remain uncertain autonomous vehicles, poll shows.)
With the introduction of the NVIDIA Ampere architecture, the Nvidia Drive platform is expanding driving capabilities from an entry-level ADAS solution all the way to a Level 5 robotaxi system.
Nvidia got started in the auto industry with its graphics and technology know-how.
According to Nvidia automakers have generally developed one computer system for ADAS systems and a different one for higher levels of automated driving, however, the development of multiple systems has become cost prohibitive.
With a single platform, Nvidia official said, developers can utilize one architecture to more easily develop autonomous driving technology across all their market segments. And since the DRIVE platform is software-defined and based on the large CUDA developer community, it can easily and constantly benefit from over-the-air updates.
(Nvidia unveils plans for higher powered processor.)
This expanded range starts at an NCAP 5-star ADAS system and runs all the way to a Drive AGX Pegasus robotaxi platform, which features two Orin SoCs and two Nvidia Ampere GPUs to 2 trillion operations per second, or — more than 6x the performance of the previous platform.