Drobot : Fast and Furious Computation for Drone Navigation

Authors: Zeryab Moussaoui, Yacine Ben Ameur , Houssem Meghnouj, Nabil Tchoulak , Merouane Guettache

23 Mars 2019: Submit the V1 code according to Alpha-Pilot deadline.

02 Avril 2019: Resolve all submission issues.

04 Avril 2019: Resolve remaining issue.

10 May 2019 : Adding technical report.

26 June 2019 : Adding Simulation of control law

The aim of this code is to complete the selection of the Lockheed Martin's Drone Race "Alpha Pilot" (https://www.herox.com/alphapilot/78-test-3) but can be used for other drone race contests.

The Alpha Pilot qualification evaluates team's skills in both Control Engineering and Real-time programming, by controlling an autonomous drone on a provided simulator, FlightGoggles

Architecture

Drobot V1 is a simple cascaded high levels position and attitude controllers, sensing pose from both IMU and camera :

Prerequirements

Recommanded Hardware

Tested on both :

• GPU 1050Ti , 16 Go of RAM
• AWS g3s.xlarge

Software

• Ubuntu 16.04.5 LTS
• ROS Kinetic
• Open CV 3.4
• CUDA drivers
• FlightGoggles

Installation

Download the repository and execute :

./install.bash && source catkin_ws/devel/setup.bash

Running

Execute ROS node :

rosrun flightgoggles scorer.sh

About the Code

• Position and Attitude controllers in Controller.py
• Position Estimation in Stereo-SLAM.cpp

Tools codes from : Drobot-dev

Technical details

Please read following article : Fast and Furious Computation for Drone Navigation

Related Publications

If you use of our code or our report in your project , please cite :

 @article{Drobot,
  title={{Drobot}: Fast and Furious Computation for Drone Navigation},
  author={Zeryab Moussaoui and al.},
  journal={Github},
  year={2019}
 }