NVIDIA’s Drive Thor Might Be the Ultimate Big Brain for Your Autonomous Car

Nvidia unveils Drive Thor, one AI chip with the aim to become the brains of self-driving car

Chip giant Nvidia Corp on Tuesday unveiled its new computing platform named DRIVE Thor that would centralize autonomously and assisted driving along with some other digital functions including in-car entertainment. Nvidia, a big tech in the gaming chip and artificial intelligence (AI) market, has made a big push in the automotive business, a key growth segment for the company.

Nvidia mentioned it designed the chip using the latest advancements in graphics and processing to provide 2,000 teraflops of performance, all while keeping costs down. Big tech is gearing up to deliver Drive Thor processors, its next-generation automotive-grade chip in 2024 hitting the roads in 2025, starting with an EV from Chinese carmaker Zeekr, stated Danny Shapiro, vice president of Nvidia's automotive work. They're based on Nvidia's new Hopper graphics processing unit to better handle the artificial intelligence software that's key to self-driving cars.

Nvidia claims that Drive Thor can unify all the various functions of vehicles — including infotainment, the digital dashboard, sensors, parking, and autonomous operation — for a wide range of efficiency. Vehicles with the chipset will be capable to run Linux, QNX, and Android simultaneously. It provides the vast processing power that autonomous vehicle operations require, automakers can even use two of the Drive Thor chipsets in tandem by employing an NVLink-C2C chip interconnect technology to have them running a single operating system.

In addition, Nvidia mentions that the SoC marks a significant leap forward in "deep neural network accuracy." The chipset has a transformer engine, fresh addition to the Nvidia GPU Tensor Core. "Transformer networks process video data as a single perception frame, enabling the compute platform to process more data over time," the big tech said. It noted that the SoC can boost inference performance of transformer deep neural networks by up to nine times, "which is paramount for supporting the massive and complex AI workloads associated with self-driving."

Thor automotive AI chip details

With 77 billion transistors, Thor will be huge, if not the biggest processor on the market. But it'll let automakers replace a heavier, more expensive, and more power-hungry collection of smaller chips, Nvidia mentions.