Nvidia’s GTC conference began with a series of announcements highlighting its role in advance of autonomous driving technology.
Chipmaker provides several Nvidia brand tools to automakers and autonomous car companies to power self-driving cars and create factory digital twins. On Tuesday, companies such as Torc, Gatik and even General Motors announced plans to use NVIDIA products in vehicle production, from warehouse robotics to autonomous driving.
To avoid confusion among the many brand names that Nvidia has awarded to hardware and software tools, here is a simple glossary of conditions.
Drive AGX: NVIDIA’s in-vehicle supercomputer that processes real-time sensor data. Drive Orin SoC: System-on-Chip, a more advanced central computer. Orin processes data from a variety of sensors and powers new AVs with level 4 autonomy. Drive Tall SOC: Thor is the next step. Optimized for trans architecture, it can accommodate generation AI and is designed to handle everything from autonomous driving operations to cockpit functions and infotainment. DriveOS: This is Nvidia’s safety-focused operating system for NVIDIA’s AV platform, which promises safe, real-time AI processing and integration of advanced driving and cockpit functions. Omniverse: Nvidia’s Omniverse is a simulation platform that allows automakers to develop and operate complex, AI-enabled virtual environments, generate synthetic data, test AV software, and build factory digital twins. Cosmos: Cosmos is built to power world model training for AVS and robotics physical AI development.
In addition to that, Nvidia announced Halos today. It is defined as AI-driven safety systems for future physical AI, such as AVS and humanoid robots. Halos has a considerable integration of Nvidia’s lineup of automotive hardware and software safety solutions, so think of it as more umbrellas.
Here’s a quick summary of Nvidia’s car announcements from the second day of the GTC.
General Motors
GM has announced that it has expanded its partnership with Nvidia with collaborations that touch on all aspects of the automaker’s business, including factories, robots and self-driving cars. Let’s start with the factory.
GM said it will use an omniverse with cosmos to train AI manufacturing models and help build next-generation factories. Omniverse allows GM to build digital twins in their factories to virtually test new production processes without disrupting, for example, existing vehicle production. We also use Omniverse to train robotics platforms for operations such as material handling and transport.
Regarding self-driving cars, GM said it will use Nvidia’s Drive AGX for in-vehicle hardware for future advanced driver assistance systems and in-cabin safety experiences.
Gatic
Gatik, an autonomous trucking company supported by Isuzu and Goodyear Ventures, is also involved in Nvidia’s automotive ecosystem. The Silicon Valley and Toronto-based company specializes in mile logistics during autonomous driving via self-driving box trucks, and says it will develop and deploy Drive Tall accelerated Drive AGX to act as an AI brain across a truck’s fleet. Gatik says it also runs AI models on its DriveOS system for safety.
The startup noted that the collaboration will help accelerate the deployment of level 4 autonomous trucks of scale for its customers, including Walmart, Kroger and Tyson Foods.
plus
Additionally, the autonomous transportation software startup said on Tuesday it will use the Cosmos World Foundation model to accelerate testing and development of its autonomous driver, SuperDrive.
According to the company, Plus’s SuperDrive system is built on top of multiple Nvidia Drive Orin Socs.
“Today, we are developing our production with Orin and working with our suppliers to produce Thor Systems (future),” Plus co-founder and CEO David Liu told TechCrunch.
The startup recently engaged in business vehicle manufacturers such as Traton Group, Iveco and Hyundai, integrating SuperDrive into the truck. Additionally, they have tested technology on public roads in Texas and Sweden, targeting commercial launches in 2027.
Torc
Yet another self-driving trucking company, Torc, announced it is working with Nvidia to develop a scalable physical AI computing system for AVS. Virginia-based Torc, a subsidiary of Daimler Truck AG, will also work with Flex, which builds an automotive-grade computing platform.
Torc says it is using cocktails from Nvidia chip architectures, including Drive AGX, Drive Olin and Drive, to support the future deployment of autonomous driving capabilities operating towards the 2027 commercial launch.
In October 2024, the company achieved its first driver-out test on the closed course in Texas.
Volvo
While Volvo is not collaborating with Nvidia to accelerate automated driving technology, the automaker relies on Nvidia’s Blackwell GPUs and relies on aerodynamic simulations. Better aerodynamics in vehicles, especially EVs, improve battery performance as the car’s resistance decreases.
Rather than using Nvidia’s Omniverse Simulator, Volvo has worked with software simulation company ANSYS to obtain a benchmark that is fast enough to run aerodynamic simulations. ANSYS’s so-called “Fluent” simulation software runs on Nvidia’s Blackwell GPU.
Volvo and ANSYS used the EX90 CAD (computer-aided design) model as data input for the Fluent simulation. This allowed Volvo to reduce multiple design iterations per day, optimized vehicle designs, and acceleration times to the market.
This article has been updated to clarify the use of ANSYS technology and Plus’ Nvidia chips.