Master project

This repository contains all the code for my master project, conducted the spring 2020. The '/catkin_ws' folder contains the developed system and is the only relevant folder. The '/investiagation' folder contains investigations and various other code that was made along the way and is kept for personal reference.

Installation

Prerequisites:

• Ubuntu Kinetic (16.04)

Install ROS Kinetic

Follow this guide: http://wiki.ros.org/kinetic/Installation/Ubuntu

Install Gazebo 7.0

sudo apt-get install ros-kinetic-gazebo-ros

ROS packages

• uuv_simulator (https://github.com/uuvsimulator/uuv_simulator) sudo apt install ros-kinetic-uuv-simulator

• ardrone_autonomy (https://ardrone-autonomy.readthedocs.io/en/latest/installation.html)

sudo apt-get install ros-kinetic-ardrone-autonomy

• tum_simulator on Kinetic and Gazebo 7 (https://github.com/eborghi10/AR.Drone-ROS) (Download files and place them in folder /ardrone_simulator_gazebo7)

Python packages

• Numpy 1.16.6
• Scipy 1.2.2
• Matplotlib 2.2.4
• OpenCV 3.3.1-dev

Can be install using pip:

sudo apt-get install python-pip
python -m pip install --user numpy==1.16.6 scipy==1.2.2 matplotlib==2.2.4

and apt-get:

sudo apt-get install python-opencv

Add the necessary models

Add models to the hidden folder .gazebo/models

• Landing platform (helipad)
• ReVolt (revolt)

Other useful things:

• Terminator, for multiple pages in one window

sudo apt-get install terminator

DDPG package

The DDPG package made by Daniel Tavakoli is added in a separate package.

To run the DDPG model:

• Tensorflow 1.15.0
• Keras 2.2.4
• tqdm 4.46.0

pip install tensorflow==1.15.0
pip install Keras==2.2.4
pip install tqdm==4.46.0

Running the system

Connect to physical quadcopter

Turn the quadcopter on and connect to it over WiFI. Then run

roslaunch uav_vision real_ar2.launch

and switch to bottom camera

rosservice call /ardrone/setcamchannel 1

Start simulator and connect to simulated quadcopter

Run

roslaunch uav_vision sim_ar2.launch

View the output from the quadcopters bottom camera

Run

roslaunch uav_vision camera_view.launch

Start the position estimator

Run

roslaunch uav_vision perception_system.launch

Joystick:

• Connect the PS4 controller to the computer via Bluetooth. Then run

roslaunch uav_vision joystick.launch

Start the PID controller

Run

rosrun uav_vision pid_controller.py

Start the Automated Landing Planner

Run

rosrun uav_vision automated_landing.py

To land with DDPG

Run

rosrun ddpg ddpg_hover_descend.py