Prerequisites for Generating Code for NVIDIA Boards

You must install required software and hardware to use NVIDIA^® Jetson™ or NVIDIA DRIVE^® boards with MATLAB^®. To use the MATLAB Coder™ Support Package for NVIDIA Jetson and NVIDIA DRIVE Platforms, you must use a supported hardware board, install required software, and set up environment variables. To target GPU devices, you can also install GPU Coder™.

Target Requirements

The target NVIDIA board must be a supported board, and you must install NVIDIA software and other libraries on the board. The support package uses environment variables to locate the required software.

Target Hardware

The MATLAB Coder Support Package for NVIDIA Jetson and NVIDIA DRIVE Platforms supports these boards:

• NVIDIA DRIVE PX 2

• NVIDIA Clara AGX™

• NVIDIA Jetson AGX Xavier™

• NVIDIA Jetson AGX Orin™

• NVIDIA Jetson Orin Nano

• NVIDIA Jetson Orin NX (GPIO workflows are not supported)

• NVIDIA Jetson Nano™

• NVIDIA Jetson TX1

• NVIDIA Jetson TX2

• NVIDIA Jetson Xavier NX

The support package uses an SSH connection over TCP/IP to execute commands while building and running the generated code on the Jetson or DRIVE platforms. Connect the target platform to the same network as the host computer. Alternatively, use an Ethernet crossover cable to connect the board directly to the host computer.

Note

On the Windows^® platform, open port 18735 in the Windows Firewall settings. This port is required to establish a connection to the MATLAB server running on the embedded platforms.

Required Target Software

Use the NVIDIA JetPack (NVIDIA) or the DriveOS SDK (NVIDIA) software to install the OS image, developer tools, and the libraries required for developing applications on the Jetson or DRIVE platforms. You can use the Component Manager in the JetPack or the Drive Platform software to select the components to install on the target hardware. For installation instructions, refer to the NVIDIA board documentation. Install these libraries on the Jetson or DRIVE board:

• CUDA^® toolkit

• NVIDIA CUDA Deep Neural Network (cuDNN) library

• NVIDIA TensorRT™ library

• OpenCV library

• GStreamer library version 1.0 or higher

The MATLAB Coder Support Package for NVIDIA Jetson and NVIDIA DRIVE Platforms supports these JetPack and DriveOS SDK versions.

Hardware Platform	Software Version
Jetson AGX Orin	JetPack 6.0 Note By default, JetPack 6 makes the GPIO pins of the board input-only. To use a GPIO pin for output, see the NVIDIA documentation to change the pinmux of the board. Generating code that uses the TensorRT library is not supported for boards configured with JetPack 6.1.
Jetson Orin Nano
Jetson Orin NX
Jetson AGX Xavier	JetPack 5.1.1
Jetson Xavier NX
Clara AGX	Holoscan SDK Note Generating code that uses the TensorRT library is not supported. Using the GPIO pins of the Clara AGX board is not supported.
Jetson Nano	JetPack 4.6.1
Jetson TX1
Jetson TX2
DRIVE	DRIVE OS 5.0.10.3-12606092

Optional Target Software

These features require additional software on the target:

• To run webcam examples, install version 1.2 of the Simple DirectMedia Layer library, the V4L2 library, and the V4L2 utilities. You must also install the development packages for these libraries.

• To deploy the Audio File Read Simulink^® block, install the Sound eXchange (SoX) utility and its development and format libraries. To install these libraries on Jetson, use the apt-get command.

  sudo -S apt-get -y install sox libsox-fmt-all libsox-dev alsa-utils

• To use Modbus^® functions and blocks, install the libmodbus libraries from https://github.com/stephane/libmodbus. To install these libraries on Jetson, use these commands:

  git clone https://github.com/stephane/libmodbus
  cd libmodbus
  ./autogen.sh
  ./configure && sudo make install
  

• To use MQTT Publish and MQTT Subscribe blocks, install the MQTT library from https://github.com/eclipse-paho/paho.mqtt.c. You must also install the json-c library.

  To install the MQTT library on Jetson, use these commands:

  git clone https://github.com/eclipse/paho.mqtt.c.git
  cd paho.mqtt.c
  make
  sudo -S make install

  To install the json-c library on Jetson, use the apt command to install the library:

  sudo apt install libjson-c-dev

Environment Variables on the Target

The support package uses environment variables to locate the tools, compilers, and libraries required for code generation. Set the environment variables in the table.

Variable Name	Default Value	Description
PATH	/usr/local/cuda/bin	Path to the CUDA toolkit executable on the Jetson or DRIVE platform.
LD_LIBRARY_PATH	/usr/local/cuda/lib64	Path to the CUDA library folder on the Jetson or DRIVE platform.

The required environment variables must be accessible from non-interactive SSH logins. For example, you can use the export command at the beginning of the $HOME/.bashrc shell configuration file to add the environment variables.

 Example .bashrc File

# ~/.bashrc: executed by bash(1) for non-login shells.
# see /usr/share/doc/bash/examples/startup-files (in the package bash-doc)
# for examples

# If not running interactively, don't do anything
case $- in
    *i*) ;;
      *) 
         export PATH=$PATH:/usr/local/cuda/bin
         export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda/lib64
         return;;
esac

# don't put duplicate lines or lines starting with space in the history.
# See bash(1) for more options
HISTCONTROL=ignoreboth

# append to the history file, don't overwrite it
shopt -s histappend

# for setting history length see HISTSIZE and HISTFILESIZE in bash(1)
HISTSIZE=1000
HISTFILESIZE=2000
.
.
.

Alternatively, you can set the system-wide environment variables in the /etc/environment file. You must have sudo privileges to edit this file.

 Example /etc/environment File

PATH="/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin"
LD_LIBRARY_PATH="/usr/local/cuda/lib64/"

Input Devices

To record video or audio, connect:

• A camera connected to the USB or CSI port of the target hardware

• A USB audio device

Development Host Requirements

The development host computer must have MATLAB and MATLAB Coder. For processor-in-the-loop (PIL) execution, you must also install a supported compiler on the host computer. Optionally, to target GPU devices or generate code for deep learning applications, install additional MathWorks^® software.

Optional MathWorks Software

• GPU Coder— Required for GPU targeting

• Parallel Computing Toolbox™— Required for GPU targeting

• Simulink— Required for generating code from Simulink models

• Simulink Coder— Required for generating code from Simulink models

• Deep Learning Toolbox™— Required for deep learning

• MATLAB Coder Interface for Deep Learning and GPU Coder Interface for Deep Learning support packages— Required for deep learning

• Embedded Coder^®— Recommended

• Computer Vision Toolbox™— Recommended

• Image Processing Toolbox™— Recommended

To verify the GPU code generation environment on the Jetson or DRIVE, use the GPU Environment Check (GPU Coder) app or the coder.checkGpuInstall (GPU Coder) function.