End-to-end solutions for autonomous vehicles

The NVIDIA DRIVE family of products for autonomous vehicle development, spanning everything from cars to data centers.

End-to-end solutions for autonomous vehicles

DRIVE Hyperion is an in-vehicle solution whose architecture includes sensors, DRIVE AGX for computing, and the software tools needed to enable powerful autonomous driving and smart cockpit capabilities. NVIDIA provides the hardware and software needed for autonomous vehicle development in the data center, including NVIDIA DGX for training DNNs for perception and DRIVE Sim for generating datasets and validating the entire autonomous driving stack.

DRIVE Hyperion

NVIDIA DRIVE Hyperion is an autonomous vehicle development platform and reference architecture for developing L2+ and L3 highway autonomous driving solutions. It consists of a complete sensor suite that has been tuned, optimized, and safety-certified, as well as the high-performance AI computing platform DRIVE AGX.

The NVIDIA DRIVE AGX Developer Kit provides the hardware, software, and sample applications needed for production-grade autonomous vehicle development. The DRIVE AGX system is built on production-grade, automotive-grade silicon, designed with safety in mind, and features an open software framework.

DRIVE Hyperion also features additional computing power to remove data recording and playback and improve overall data processing efficiency.

This end-to-end (E2E) platform can be integrated into test vehicles, allowing developers to build, evaluate and validate autonomous driving technology at scale.

DRIVE SDK

NVIDIA DRIVE SDK includes the basic DRIVE OS and DriveWorks SDK, as well as advanced applications such as highly automated supervised driving (DRIVE AV) and AI cockpit (DRIVEIX). It is a modular, open platform that helps developers more efficiently build and deploy a variety of advanced autonomous driving functions, including perception, positioning and mapping, planning and control, driver monitoring, and natural language processing.

DRIVE Sim

NVIDIA DRIVE Sim uses high-fidelity and physically accurate simulation technology to create a safe, scalable and cost-effective way to deploy autonomous vehicles. It uses NVIDIA core technologies such as NVIDIA RTX, Omniverse and AI to create a powerful cloud computing platform that can generate many real-world scenarios for autonomous driving development and verification. DRIVE Replicator can generate data sets to train the vehicle's perception system or test the decision-making process. It can also support connection to any autonomous driving stack in a "software-in-the-loop (SIL)" or "hardware-in-the-loop (HIL)" configuration, using DRIVE Console test system integration.

NVIDIA DGX

NVIDIA DGX is a single-box supercomputer that provides you with the computing power you need to accelerate the training of deep neural networks (DNNs) for autonomous driving perception or run non-stop inference (replay) workloads for validation. It is used as a core building block for complete systems in AI data centers, helping to improve developer productivity and accelerate iterations for training large DNNs.