At the recent IAA Mobility conference in Munich, NVIDIA outlined how cloud-to-car AI platforms are bringing new levels of safety, intelligence, and trust to the road.
NVIDIA and its partners didn’t just show off cars at the conference — they showed off what cars are becoming: AI-defined machines, built as much in the data centre as they are in the factory.
Vice President of Automotive Ali Kani explained: “Vehicles are moving from being dependent on horsepower to compute power, from mechanical systems to software stacks.
“Automotive engineering is now infused with silicon acceleration, as automakers and suppliers adopt NVIDIA’s cloud-to-car platform to drive safety, intelligence and efficiency into tomorrow’s vehicles.”
NVIDIA claims to offer an end-to-end compute stack for autonomous driving. Its three AI compute platforms are: NVIDIA DGX for training AI in data centres, NVIDIA Omniverse and Cosmos to simulate worlds and generate synthetic data for testing and validation, and NVIDIA DRIVE AGX in-vehicle computers to process sensor data in real time.
Together, these platforms are said to form a feedback loop for learning, testing and deployment that tightens the cycle of innovation while keeping safety front and centre.
All About Safety
NVIDIA Halos is a full-stack, comprehensive safety system that integrates hardware, software, AI models, and tools to ensure safe AV development and deployment from cloud to car. It unifies vehicle architecture, AI models, chips, software, tools, and services to ensure the safe development of autonomous vehicles, from cloud to car.
NVIDIA Halos brings together safety-assessed systems-on-a-chip, the safety-certified NVIDIA DriveOS operating system, and the DRIVE AGX Hyperion architecture into a unified platform for autonomous driving. This platform is backed by the NVIDIA Halos Certified Program and its AI Systems Inspection Lab, which deliver rigorous validation to ensure real-time AI operates with end-to-end reliability.
Kani added: “With AI-driven workflows and high-fidelity sensor simulations built with NVIDIA Omniverse and Cosmos, automakers can train, test and safely validate vehicle performance – even under conditions that are hard to experiment with in the real world, such as rare or hazardous traffic situations and edge-case events, and in complex environments.”
Simulation tools are increasingly critical for advancing safe, scalable autonomous vehicle development.
CARLA is one of the world’s most popular open-source simulation platforms with more than 150,000 active developers, serving as a testbed for AV research and development. NVIDIA is partnering with CARLA to integrate its latest NuRec rendering APIs and Cosmos Transfer world foundation model. This enables developers to generate sensor data from Gaussian representations with ray tracing and amplify diversity with Cosmos WFMs.
Capgemini and TCS are already tapping into this integration to expand their simulation capabilities and push the boundaries of software-defined vehicle development.
Expanding the Ecosystem
Automotive manufacturers such as incumbents Volvo, Mercedes-Benz, Lotus, as well as electric newcomers Lucid, ZYT and XPENG are using NVIDIA’s cloud-to-car AI platform to transform their next-generation vehicles.
Beyond automakers, technology companies including MediaTek, ThunderSoft, Cerence, ZF Group, RoboSense, and Magna are building on NVIDIA AI to accelerate the development of software-defined vehicles. These advances include intelligent cockpits, voice interaction, advanced driver-assistance systems, automated driving, chassis control, and full autonomous solutions.
