Nvidia unveils Drive Thor, one chip to rule all software-defined vehicles • TechCrunch


Nvidia is gearing up to unveil Drive Thor, its next-generation automotive-grade chip, which the company claims will be able to unify a variety of in-car tech from self-driving features and driver monitoring systems to streaming Netflix for kids.

The Thor, due to enter production in 2025, isn’t just notable because it’s a step up from Nvidia’s Drive Orin chip. It also occupies Drive Atlan’s place in the lineup.

Nvidia is scrapping the Drive Atlan SoC for Thor ahead of schedule, Founder and CEO Jen-Hsun Huang said Tuesday at the company’s GTC event. In the race to develop a bigger and better chip, Nvidia chose Thor, which has 2,000 teraflops of performance and will provide twice the compute and throughput, according to the company.

“If we look at a car today, advanced driver assistance systems, parking, driver monitoring, camera mirrors, digital The instrument cluster and infotainment system are both different computers distributed throughout the vehicle,” Monday’s briefing. “By 2025, these functions will no longer be separate computers. Instead, Drive Thor will enable manufacturers to efficiently incorporate these functions into a single system, reducing overall system cost.”

One chip to rule them all. A chip that helps automakers build software-defined self-driving cars. A chip can be continuously upgraded over the air.

Nvidia already has several automotive customers using Drive chips to build software-defined fleets. Volvo, for example, announced at its annual CES tech conference in January that its new self-driving features will be powered by Drive Orin. The automaker also said it will use Qualcomm’s Snapdragon chip to power its infotainment system. It’s this space-sharing with rivals that could prompt Nvidia to create more powerful chips.

Zeekr was the first to raise his hand for Thor. The Chinese luxury electric vehicle startup, owned by Geely, said it will use the advanced chip in its next-generation vehicles starting in 2025, according to Shapiro.

The Xpeng G9 SUV is already using the latest-generation chip Drive Orin, which will be able to support highly advanced driver assistance functions, “such as parking and driving on primary and secondary streets, highways and private roads, as well as safe handling, entering and exiting highways. , urban byways and toll routes,” Shapiro said. Undoubtedly, Xpeng, which recently launched the city-navigation navigation ADAS in its P5 sedan and plans to launch in the G9, will sign the upgraded chip.

Shapiro also noted that self-driving solutions provider QCraft will start a self-driving taxi business in China, powered by Orin.

Other automakers that have previously announced the use of Nvidia Drive Orin include Baidu’s electric vehicle company JiDU Auto, NIO, Li Auto, R Auto, IM Motors and Polestar. It’s worth noting that quite a few of Nvidia’s automotive customers are located in China. While the chipmaker is headquartered in California, its chips, along with those of nearly every other company, are made in Taiwan.

Can Nvidia deliver Raytheon to Chinese customers?

Earlier this month, the U.S. government imposed export restrictions on advanced artificial intelligence chips from China, including Hong Kong and Russia.Nvidia not selling to Russia, but China sanctions could Cost the company up to $400 million Potential sales for the third quarter. The U.S. said the move would address the risk of chips being used or diverted to “military end-use” or “military end-users” in China and Russia, but it was also part of a Biden administration’s effort to prevent China from becoming an essential and lucrative tool for dominate the chip production industry.

The government specifically restricted access to Nvidia’s A100 and H100 graphics processing units. Fortunately for Nvidia, the company can continue to manufacture the H100 in China, although purchases by Chinese customers will be limited.

Shapiro said automotive customers will not be affected by the restrictions on Nvidia’s high-end data center products, and the company is working with Chinese customers and the U.S. government “to come up with different alternatives that are not bound by the same license requirements.”

Updates to Drive SIM and Digital Twin Technology

Heatmap in Simulation

Nvidia Drive SIM is built on Nvidia Omniverse and provides the core simulation and rendering engine for autonomous vehicle development.

At the GTC event, Nvidia also announced that its end-to-end simulation platform, Drive SIM, is getting a new set of artificial intelligence tools, called a “Neural Reconstruction Engine,” to assist in the testing and development of autonomous vehicles.

“Using a neural engine consisting of multiple [deep neural networks]real-world rides can be accurately recreated and played back in the simulation, with the ability to change sensor configurations and locations, create new scenarios, and modify or add new aspects of other road user behavior,” Shapiro said.

The way it works is that AI can deconstruct a 3D scene from recorded sensor data, which can then be enhanced in Drive SIM with human-created content or AI-generated content.

Nvidia said the upgrades will also allow car designers, software engineers and electronics engineers to collaborate in Drive SIM to simulate the software inside the car.

The company also announced the second generation of Nvidia OVX, which will provide an immersive urban digital twin that can run in Nvidia’s Omniverse.

BMW Group and Jaguar Land Rover were among the first clients to use OVX, Shapiro said, noting that marketing group WPP is using the Omniverse cloud to create a suite of services for automotive clients, such as personalised programmatic advertising, with “perfect photo-real” The content is “through virtual sets, which will save car companies orders of magnitude costs” for expensive photo and video shoots, often in scenic locations around the world.



Source link

Leave a Reply

Your email address will not be published.