CSE 326 - Project-Based Learning

An Integrated Framework for Autonomous Navigation and Machine Vision in Micro Aerial Vehicles

Group: F2022 - 05

Contributors:

Name	ID
Ahmed Abdelkader	120200028
Ahmad Mongy	120200033
Mahmoud Akrm	120200045
Mohamed Ayman	120200081
Mohanned Ahmed	120200113
Peter Fayez	120200073
Yasser Ossama	120190122
Ziad Hesham	120200078

Native Environemnt Setup

Setting up the simulation environment from scratch requires some effort, but results in the most performant setup, with less chance of driver issues.

Prerequisites: Ubuntu 20.04.

Install ROS Noetic using the official installation manual (Desktop or Full install).

Add sourcing ROS' setup.bash initialization script to your .bashrc:

echo "source /opt/ros/noetic/setup.bash" >> ~/.bashrc
source ~/.bashrc

Install required tools:

sudo apt install build-essential git python3-pip python3-rosdep

Install required tools:

sudo apt install build-essential git python3-pip python3-rosdep

Create a workspace for the simulation

Create a workspace for the simulation:

mkdir -p ~/catkin_ws/src
cd ~/catkin_ws
catkin_make
echo "source ~/catkin_ws/devel/setup.bash" >> ~/.bashrc
source ~/.bashrc

Clone Clover sources:

cd ~/catkin_ws/src
git clone --depth 1 https://github.com/CopterExpress/clover
git clone --depth 1 https://github.com/CopterExpress/ros_led
git clone --depth 1 https://github.com/ethz-asl/mav_comm

Install all dependencies using rosdep:

cd ~/catkin_ws
sudo rosdep init
rosdep update
rosdep install --from-paths src --ignore-src -y

Install Python dependencies:

sudo /usr/bin/python3 -m pip install -r ~/catkin_ws/src/clover/clover/requirements.txt

Get PX4 sources

PX4 will be built along with the other packages in our workspace. You may clone it directly into the workspace or put it somewhere and symlink to ~/catkin_ws/src. We will need to put its sitl_gazebo and mavlink submodules into ~/catkin_ws/src as well.

Clone PX4 sources and make the required symlinks:

git clone --recursive --depth 1 --branch v1.12.3 https://github.com/PX4/PX4-Autopilot.git ~/PX4-Autopilot
ln -s ~/PX4-Autopilot ~/catkin_ws/src/
ln -s ~/PX4-Autopilot/Tools/sitl_gazebo ~/catkin_ws/src/
ln -s ~/PX4-Autopilot/mavlink ~/catkin_ws/src/

Install PX4 prerequisites

PX4 comes with its own script for dependency installation. We may as well leverage it:

cd ~/catkin_ws/src/PX4-Autopilot/Tools/setup
sudo ./ubuntu.sh

This will install everything required to build PX4 and its SITL environment.

Install more required Python packages:

pip3 install --user toml

Add the Clover airframe

Add the Clover airframe to PX4 using the command:

ln -s ~/catkin_ws/src/clover/clover_simulation/airframes/* ~/PX4-Autopilot/ROMFS/px4fmu_common/init.d-posix/airframes/

Install geographiclib datasets

mavros package requires geographiclib datasets to be present:

sudo /opt/ros/noetic/lib/mavros/install_geographiclib_datasets.sh

Install other requirements

check installed packages against requirments.txt and install any missing ones.

Update Workspace

Download and extract the repo in the home directory. In case prompted, confirm replacing already existing files.

Build the simulator

Build your workspace:

cd ~/catkin_ws
catkin_make

Run the simulator

In order to be sure that everything was built correctly, try running the simulator for the first time:

roslaunch clover_simulation simulator.launch

Changing Simulation Worlds

In simulator.launch you can choose which world file to load for simulation; however, note that a complex world will require more resources to run smoothly.