GTC 2025 Highlights: NVIDIA, GM, Ansys, Volvo, Plus, Nexar, Gatik, Uber, and Tor - A summary of collaborations and advancements in self-driving technology and autonomous vehicles featuring these key players.
In a significant move to advance autonomous vehicle (AV) technology, NVIDIA has introduced Halos, a comprehensive safety framework designed to ensure the development of safe and explainable AI-driven AV systems.
Halos unifies vehicle architecture, AI models, chips, software, tools, and services, providing a full-stack safety framework from cloud to car. It incorporates design-time, deployment-time, and validation-time guardrails using NVIDIA's DGX for AI training, Omniverse with Cosmos for simulation and validation, and DRIVE AGX for deployment. This setup enables continuous safety assurance, regulatory compliance, and enhanced system reliability.
For partners in the autonomous transportation sector, Halos offers a platform that supports a closed-loop development process. Car manufacturers and developers can build, test, and validate advanced self-driving systems more effectively. Halos addresses critical industry challenges by providing explainability and handling rare, complex scenarios, thereby fostering trust and accelerating the rollout of safe AV technology.
One such partner is Plus, which is integrating NVIDIA CosmosTM world foundation models into its SuperDriveTM virtual driver system, accelerating the development of Level 4 self-driving trucks. Other companies like Gatik, Uber Freight, Torc, Magna, and Nexar are also leveraging NVIDIA's technology. Gatik, for instance, is using NVIDIA DRIVE AGX for AI processing in its autonomous vehicles, while Magna is developing driving systems using NVIDIA DRIVE Thor, based on the high-performance NVIDIA Blackwell architecture.
Nexar's partnership with NVIDIA aims to strengthen NVIDIA Cosmos world foundation models and accelerate Nexar's AI model training through NVIDIA DGX Cloud. This collaboration enhances AV training, high-definition mapping, and predictive modeling.
The integration of NVIDIA's Omniverse Blueprint with CARLA, an open-source AV simulator, and Capgemini's adoption of its capabilities signal a major leap forward in developing and validating AV technologies.
Halos also includes the AI Systems Inspection Lab, allowing partners to verify the safe integration of their products within the ecosystem, enhancing collaboration and innovation in this rapidly evolving market.
In essence, NVIDIA Halos acts as both a technical and regulatory enabler, helping partners meet stringent safety standards while advancing the scalability and reliability of autonomous and intelligent mobility solutions. As the autonomous transportation sector grows and diversifies, Halos is poised to play a crucial role in ensuring the safe deployment of these technologies.
The finance industry could benefit from NVIDIA's Halos due to its capability to ensure regulatory compliance and enhance system reliability in the autonomous transportation sector. As technology advances, transportation companies like Gatik and Magna are leveraging Halos for AI processing and driving system development, thereby improving the safety and efficiency of their autonomous vehicles (AVs). In the realm of technology, Halos fosters trust by providing explainability and handling rare, complex scenarios, which could potentially spur innovation and accelerate the rollout of safe AV technology across industries.
