To connect and work with your RC throughout the rest of the project, you'll need two things:
If you're using Linux or a Mac, you're all set. They come with a SSH client pre-installed, and you just need to open up a terminal and type:
ssh username@ipAddress
If you're using Windows, you need to install one. I'd recommend using MobaXTerm:
Download the installer or the portable version and install/unpack it
To SSH to a device:
Open MobaXTerm
Press on the Start local terminal button
ssh username@ipAddress
If your RC is connected to your network via an ethernet cable, you should be able to find your IP address:
Through your Router/Gateway interface, by looking at the DHCP leases;
Or by connecting your Nano to a monitor, keyboard and mouse, opening up a terminal and writing:
ip addr show
I would highly recommend this approach, so you can connect your RC to a WiFi network and take it and your laptop with you and connect to it on the fly.
To connect it to a WiFi network, you either need to first SSH into it over ethernet, or connect it to a monitor, keyboard and mouse, and do the following:
Connect to a WiFi network:
nmcli device wifi connect YOUR-SSID password YOUR-PASSWORD
Or make the Jetson into a hotspot so you can connect your laptop to it:
nmcli dev wifi hotspot ifname wlan0 ssid HOTSPOT-SSID password HOTSPOT-PASSWORD
What I like to do is connect it to both my home network, and a hotspot on my mobile phone, so I can use it anywhere and still have Internet access on both it and my laptop.
To do so, I use the nmcli autoconnect.priority property, so my home network has a higher priority than my phone hotspot, in case I forget to turn it off while I'm at home, so it doesn't eat up my data plan.
You can find all of your network connections saved in /etc/NetworkManager/system-connections/, which you can open up with a text editor and edit the autoconnect.priority property for each network. The higher the integer you assign to it, the higher the priority.
As an example, the network connection profile for my hotspot looks something like:
[connection]
id=Hotspot
uuid=random-long-string
type=wifi
autoconnect_priority=2 # The home network has a priority of 3, in my case
permissions=
If your Nano keeps dropping the connection for some reason, try disabling the power saving mode found in /etc/NetworkManager/conf.d, using a text editor.
Also, I assumed your Nano already has the nmcli or the NetworkManager utility installed, since it, at the time of writing, comes pre-installed with any Ubuntu distro. If, for some reason, you don't have it, you can install it using sudo apt install network-manager.
After connecting your Nano to a WiFi network you want, find out its IP Address by opening up a terminal and typing:
ip addr show
Open up a terminal on your Nano and install the following dependencies:
sudo apt-get update
sudo apt-get upgrade
sudo apt-get install build-essential python3 python3-dev python3-pip libhdf5-serial-dev hdf5-tools nano ntp
# Install the venv package
pip3 install virtualenv
python3 -m virtualenv -p python3 env --system-site-packages
# Activate the venv automatically at boot
echo "source env/bin/activate" >> ~/.bashrc
source ~/.bashrc
First, since OpenCV needs more than 4GB of RAM to be built from source, and our Jetson Nano just doesn't have that much RAM, we have to define some swap space to prevent it from going bonkers while compiling it:
# Allocates 4G of additional swap space at /var/swapfile
sudo fallocate -l 4G /var/swapfile
# Permissions
sudo chmod 600 /var/swapfile
# Make swap space
sudo mkswap /var/swapfile
# Turn on swap
sudo swapon /var/swapfile
# Automount swap space on reboot
sudo bash -c 'echo "/var/swapfile swap swap defaults 0 0" >> /etc/fstab'
# Reboot
sudo reboot
Now, we need to get all the prerequisites needed to build OpenCV from source:
# Update
sudo apt-get update
sudo apt-get upgrade
# Pre-requisites
sudo apt-get install build-essential cmake unzip pkg-config
sudo apt-get install libjpeg-dev libpng-dev libtiff-dev
sudo apt-get install libavcodec-dev libavformat-dev libswscale-dev libv4l-dev
sudo apt-get install libxvidcore-dev libx264-dev
sudo apt-get install libgtk-3-dev
sudo apt-get install libatlas-base-dev gfortran
sudo apt-get install python3-dev
Okay, let's download the source code for OpenCV which we'll be building it from:
# Create a directory for opencv
mkdir -p projects/cv2
cd projects/cv2
# Download sources
wget -O opencv.zip https://github.com/opencv/opencv/archive/4.1.0.zip
wget -O opencv_contrib.zip https://github.com/opencv/opencv_contrib/archive/4.1.0.zip
# Unzip
unzip opencv.zip
unzip opencv_contrib.zip
# Rename
mv opencv-4.1.0 opencv
mv opencv_contrib-4.1.0 opencv_contrib
Also we'll need numpy in our virtual environment for this to work:
# Install Numpy
pip install numpy
We also need to make sure CMake correctly generates the OpenCV bindings for our virtual environment:
# Create a build directory
cd projects/cv2/opencv
mkdir build
cd build
# Setup CMake
cmake -D CMAKE_BUILD_TYPE=RELEASE \
-D CMAKE_INSTALL_PREFIX=/usr/local \
-D INSTALL_PYTHON_EXAMPLES=ON \
-D INSTALL_C_EXAMPLES=OFF \
-D OPENCV_ENABLE_NONFREE=ON \
# Contrib path
-D OPENCV_EXTRA_MODULES_PATH=~/projects/cv2/opencv_contrib/modules \
# Your virtual environment's Python executable
# You need to specify the result of echo $(which python)
-D PYTHON_EXECUTABLE=~/env/bin/python \
-D BUILD_EXAMPLES=ON ../opencv
The cmake command shows a summary of its configuration, and you should make sure that the Interpreter is set to the Python executable of your virtual environment, not the base OS one.
Now, to compile the code from the build folder, run the following:
make -j2
# Install OpenCV
sudo make install
sudo ldconfig
This will take a while. And by a while, I mean: Go grab a cup of coffee and watch a TV Show or a movie or something while.
Now we just need to link it to our virtual environment:
cd to: /usr/local/lib/python[YOUR.VERSION]/site-packages/cv2/python[YOUR.VERSION] and do ls to find out the exact name of the .so we built.
It should look something like: cv2.cpython-[YOURVERSION]m-[***]-linux-gnu.so
Rename it to cv2.so: mv cv2.cpython-whatever-the-full-name-is.so cv2.so
And finally:
# Go to your virtual environments site-packages folder
cd ~/env/lib/python[YOUR.VERSION]/site-packages/
# Symlink the native library
ln -s /usr/local/lib/python[YOUR.VERSION]/site-packages/cv2/python-[YOUR.VERSION]/cv2.so cv2.so
To make sure everything works as it should, run:
import cv2
# Should print 4.1.0
print(cv2.__version__)
First, go to a directory where you'd like your stuff to be:
# Probably
cd ~/projects
Install the latest Donkey from GitHub:
# Clone it from GitHub
git clone https://github.com/autorope/donkeycar
cd donkeycar
# Checkout the master branch
git checkout master
pip install -e .[nano]
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v42 tensorflow-gpu==1.13.1+nv19.3