Nvidia Unveils Drive AGX Orin Smart Chipset For Autonomous Cars, Robots

Nvidia is a well-known brand when it comes to computer graphics, having launched more than a dozen GPUs just in 2019. Now, the company has announced its latest innovation -- the Nvidia Drive AGX Orin, which is a chipset for autonomous cars and robots.

The Nvidia Drive AGX Orin is a platform built on top of an SoC (system on chip) named Orin. The chipset consists of more than 17 billion transistors and is based on Nvidia's Arm Hercules CPU cores and next-gen Nvidia GPU. Nvidia claims that the chipset can deliver 200 trillion operations per second, which is 7 times more than the Nvidia Xavier SoC.

The chipset is designed to handle a large number of applications using deep neural networks, especially on autonomous vehicles and robots. The chipset is also ISO 26262 ASIL-D certified, which is a safety standard for autonomous vehicles. CPU, GPU, and AI will work together to make the necessary calculations in real-time. As of now, there is no information if the chipset requires an active internet connection or not.

This chipset can be used on vehicles from level-2 self-driving to level-5 self-driving, depending on the requirement of OEMs. As the chipset is based on open CUDA and TensorRT APIs and libraries, engineers can get the most out of these chipsets for a long time.

Availability Of Nvidia Drive AGX Orin

The Nvidia Drive AGX Orin will be available in various configurations based on a single architecture to auto and robot markers from 2022. This means we can expect to see final products powered by Nvidia Drive AGX Orin probably by 2023.

Creating a safe autonomous vehicle is perhaps society's greatest computing challenge. 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, says Jensen Huang, founder, and CEO of Nvidia