Overview

The NVIDIA DRIVE® OS 6.0 Linux Software Development Kit enables application development with NVIDIA DRIVE Orin SoC for Automotive. At a high level, the SDK consists of:
  • DRIVE OS foundation components including bootloaders, a Type 1 hypervisor, and virtualization.
  • GuestOS virtual machines including Linux operating systems for system software and application development.

The DRIVE OS Software Development Kit (SDK) provides industry standard APIs including NvMedia, CUDA and TensorRT. The CUDA and TensorRT modules included in DRIVE OS 6.0 software releases are compatible only with the Automotive DRIVE AGX Orin Platform. As such, they must only be used with DRIVE OS. These modules must not be used standalone as they are not compatible with other NVIDIA devices.

What is NVIDIA DRIVE OS 6.0?

NVIDIA DRIVE® OS 6.0 is the reference operating system and associated software stack designed specifically for developing and deploying autonomous vehicle applications on DRIVE Orin-based hardware. DRIVE OS 6.0 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 NVIDIA® CUDA® libraries, NVIDIA TensorRT, NvMedia, and other components optimized to provide direct access to DRIVE AGX Orin hardware acceleration engines.

What is NVIDIA DRIVE OS 6.0 SDK?

DRIVE OS 6.0 SDK consists of DRIVE Orin software, libraries, and tools to build, debug, profile, and deploy applications for autonomous vehicles and self-driving cars across the CPU, GPU and other DRIVE AGX Orin hardware acceleration engines. These development tools provide optimized workflows for parallel computing and deep learning development.

In order to maximize productivity, DRIVE OS 6.0 SDK leverages industry standard tools, technologies, and APIs to provide a familiar and comfortable high-productivity development environment.

Getting Started with NVIDIA DRIVE OS SDK

Topic 

Where to Learn More 

Installation

See Installation for information on installing or upgrading a DRIVE OS SDK installation on a host machine.

Flashing

See Flashing DRIVE AGX Orin for information on updating the system image and associated firmware on a DRIVE AGX Orin system.

Architecture

See DRIVE OS Architectural Overview for an architectural overview of the DRIVE OS stack.

What is NVIDIA DRIVE OS 6.0 PDK?

NVIDIA DRIVE® OS 6.0 Platform Development Kit (PDK) is used to modify DRIVE OS 6.0 to run on non-reference DRIVE AGX Orin-based hardware platforms. The NVIDIA DRIVE OS PDK requires specific agreements with NVIDIA. Consult with your NVIDIA Customer Support Engineer for more information.

Additional Platform Components

In addition to the Foundation services components and the DRIVE OS-specific components, additional components are available. These components are provided separately for customizing the platform development and include:

Components Description
CUDA

CUDA® is a parallel computing platform and programming model developed by NVIDIA for general computing on graphical processing units (GPUs). With CUDA, developers are able to dramatically speed up computing applications by harnessing the power of GPUs.

Consult the CUDA Samples provided as an educational resource.

Consult the CUDA Computing Platform Development Guide for general purpose computing development.

cuDNN

The NVIDIA CUDA® Deep Neural Network library (cuDNN) is a GPU-accelerated library of primitives for deep neural networks. cuDNN provides highly tuned implementations for standard routines such as forward and backward convolution, pooling, normalization, and activation layers. cuDNN is part of the NVIDIA Deep Learning SDK.

Consult the cuDNN Deep Neural Network Library of primitives for deep neural network development.

TensorRT

NVIDIA TensorRT™ is a high-performance deep learning inference optimizer and runtime that delivers low latency, high-throughput inference for deep learning applications.

Consult the TensorRT Documentation for deep learning development.