NVIDIA Tegra
NVIDIA DRIVE OS 5.1 Linux

Development Guide
5.1.3.0 Release


 
What is DRIVE OS?
 
DRIVE OS Stack Architecture
Installing DRIVE OS
Develop Applications
DRIVE OS provides an end-to-end development platform software and hardware stack for developing autonomous vehicles. Use DRIVE OS to design and develop In-Vehicle Infotainment (IVI) to turn your vehicle into an autonomous perception learning machine.
Additionally, DRIVE OS provides design flexibility and ease of integration; wherein the applications can be used directly as plug-n-play components, or customized and enhanced with designs that meet specialized product requirements and use cases.
Note:
5.1.3.0 is the last release to support Ubuntu 16.04 for host development. 5.1.6.0 will require Ubuntu 18.04 for host development.
DRIVE OS Stack Architecture
Use the DRIVE OS software stack architecture to build your autonomous vehicle applications. The architecture stack is as follows:
A screenshot of a video game Description automatically generated
Installing DRIVE OS
Installation of the DRIVE OS must be performed using the SDK Manager automated installer to step through the installation and setup of the platform on the host and target system.
Consult the SDK Manager Quickstart Guide for detailed instructions for installing the DRIVE OS.
The SDK Manager interface must be used to flash the system with the desired configuration.
Develop Applications
After installing and flashing the platform, you are ready to setup the host and target configuration.
Once the host and target system are properly configured, you are ready develop applications for:
Component
Where to Learn More
NvMedia
Consult building and running NvMedia sample applications to build an NvMedia sample application.
CUDA
Consult the CUDA Samples provided as an educational resource.
Consult the CUDA Computing Platform Development Guide for general purpose computing development.
cuDNN
Consult the cuDNN Deep Neural Network Library of primitives for deep neural network development.
TensorRT
Consult the TensorRT Documentation for deep learning development.