Introduction

Scalable AI Platform for Autonomous Driving

Autonomous vehicles will transform the way we live, work, and play, creating safer and more efficient roads. To realize these revolutionary benefits, the car of the future will require a massive amount of computational horsepower. Tapping into decades-long experience in AI, NVIDIA DRIVE® hardware and software solutions deliver industry-leading performance to help automakers, truck makers, tier 1 suppliers, and startups make autonomous driving a reality.

NVIDIA DRIVE platforms include an in-vehicle computer (NVIDIA DRIVE AGX) and complete reference architecture (NVIDIA DRIVE Hyperion™), as well as data center-hosted simulation (NVIDIA DRIVE Constellation™) and deep neural network (DNN) training platforms (DGX™). These platforms also include rich software developer kits (SDKs) to accelerate autonomous vehicle (AV) development.

To learn more about the NVIDIA DRIVE platform, see:

https://developer.nvidia.com/drive

The NVIDIA DRIVE AGX Orin™ Developer Kit provides the hardware, software, and sample applications needed for development of production level autonomous vehicles (AV). The NVIDIA DRIVE AGX System is built on production auto-grade silicon, features an open software framework, and has a large ecosystem of automotive partners (include auto-grade sensor vendors, automotive Tier 1 suppliers) to choose from.

The DRIVE AGX development environment requires an Ubuntu host development PC (not included with the DRIVE AGX Developer Kit). The DRIVE OS SDK can be installed on the host PC using either NVIDIA SDK Manager or NVIDIA GPU Cloud (NGC).

To learn more about the DRIVE AGX Orin Developer Kit, see:

https://developer.nvidia.com/drive/drive-hyperion#section-1

NVIDIA DRIVE® OS is the reference operating system and associated software stack designed specifically for developing and deploying autonomous applications on DRIVE AGX-based hardware. NVIDIA DRIVE OS delivers a safe and secure execution environment for safety-critical applications, providing services such as secure boot, security services, firewall, and over-the-air updates. The included foundational software stack consists of a Type-1 Hypervisor, NVIDIA® CUDA® libraries, NVIDIA TensorRT™, NvMedia, and other components optimized to provide direct access to DRIVE AGX hardware acceleration engines.

NVIDIA® DriveWorks SDK is middleware for autonomous vehicle software development. Included in each release are a collection of modules for common AV software use cases, such as interfacing with the vehicle’s sensors, processing sensor data to be used as input to perception algorithms, and performing DNN inference on pre-trained networks. Also included are a set of tools that support AV software development, such as sensor data recording, sensor calibration, and DNN optimization. DriveWorks abstracts away the details of the underlying DRIVE AGX platform while still exposing its power, thereby reducing the time to develop complex and accelerated AV applications.

To learn more about DriveWorks, see:

https://developer.nvidia.com/drive/driveworks

NVIDIA SDK Manager provides an end-to-end development environment setup solution for NVIDIA's DRIVE, Jetson, Clara Holoscan, Rivermax, DOCA and Ethernet Switch SDKs for both host and target devices.

To learn more about NVIDIA SDK Manager, see:

https://developer.nvidia.com/nvidia-sdk-manager

NVIDIA NGC™ hosts a catalog of GPU-optimized software for AI practitioners to develop their AI solutions. It also provides access to various AI services including NVIDIA Base Command for model training, NVIDIA Fleet Command to deploy and monitor models, and the NGC Private Registry for securely accessing and managing proprietary AI software. Also, NVIDIA AI Enterprise customers can request support through the NGC portal.

To learn more about NVIDIA GPU Cloud, see:

https://www.nvidia.com/en-us/gpu-cloud/