NVIDIA recently launched NVIDIA DRIVE AGX Orin, a highly developed software-defined platform that serves all requirements of autonomous robots and vehicles. The new platform extracts its power from new system-on-chip (SoC), known as Orin, and is made up of 17 billion transistors.
The product is based on four years of R&D Investment of NVIDIA. Orin SoC brings together NVIDIA’s next-gen GPU architecture and Arm Hercules CPU cores along with new computer vision architecture and deep learning. The NVIDIA DRIVE AGX Orin is capable of delivering 200 trillion operations per second, which is approximately seven times better than the performance of NVIDIA’s previous generation Xavier SoC.
What’s special about it?
NVIDIA’s Orin is designed to take care of a large number of applications and deep neural networks that simultaneously run in autonomous robots and vehicles. They are also capable of achieving benchmark safety standards like the ISO 26262 ASIL-D.
The DRIVE AGX Orin enables most architecturally compatible platforms that scale from level two to five in vehicles. It also enables OEMs to offer full-scale and complicated families of several software products. As both Xavier and Orin can be programmed via TensorRT APIs and open CUDA and libraries, developers will be able to leverage all of their investments through multiple product generations.
“Creating a safe autonomous vehicle is perhaps society’s greatest computing challenge,” said Jensen Huang, the CEO and founder of NVIDIA. “The amount of investment required to deliver autonomous vehicles has grown exponentially and the complexity of the task requires a scalable, programmable, software-defined AI-platform like Orin.”