Accelerating Autonomous Vehicle Development With AI

The race to develop self-driving cars requires three distinct yet crucial parts operating simultaneously: AI training, simulation, and autonomous driving—each requiring its own accelerated computing platform. This talk will explore how a model is first trained on an AI supercomputer like NVIDIA DGX, then tested and validated in simulation—using NVIDIA Omniverse running on an NVIDIA OVX system—before entering the vehicle. There, the NVIDIA DRIVE AGX platform processes sensor data through the model in real time, using NVIDIA DriveOS—the operating system for safe, AI-defined autonomous vehicles. Together, these full-stack systems that are purpose-built for these steps form a powerful trifecta that enables continuous development cycles, ultimately improving performance and delivering greater safety on our roads.