Web-based Ground Control Station

The goal of this project is to develop a cloud-based ground control station for Ardupilot drones. The web app is designed to be easily scalable so users can control multiple drones at the same time.

The project is still under development and we would love to have more developers joining us!

System architecture:

Our website

A deployed CloudStation

Deployment instructions

Prerequisite

Python 3.6+
Django
Django Channels
Django Background Tasks
Redis
pyMavlink
pyserial
Docker

Tutorial to setup Python Django development environment: https://developer.mozilla.org/en-US/docs/Learn/Server-side/Django/development_environment

Getting started

These steps can be used to set up a local development environment for this web app. Please follow deployment instructions instead if you want to deploy it on a web server.

1. Install libraries

   pip3 install -r requirements.txt  
   

2. Install and run docker: https://www.docker.com/get-started

   docker run -p 6379:6379 -d redis:2.8
   

3. Register for MapBox and generate a public access token https://account.mapbox.com/
4. Add mapbox public access token (public key) to .env of your environment Note: Please do not push your pk to GitHub. In production env, this key should ideally be temporary and change from session to session.
   • ubuntu
     export MAPBOX_PUBLIC_KEY=pk.xxxxxxxxxxxxxxxxxxxxxxxxxx
   • windows
     setx MAPBOX_PUBLIC_KEY "pk.xxxxxxxxxxxxxxxxxxxxxxxxxxxxx" or add an entry in system variables

Run server

python3 manage.py makemigrations
python3 manage.py migrate
python3 manage.py runserver
python3 manage.py process_tasks

Architecture

Milestones & Backlogs

• System architecture and prototype (8/31/19)
• Rewrite mavlink streaming code (9/5/19)
• Add map to html (10/15/19)
• Deploy on AWS (10/15/19) We are online!!!
• Build hardware stack with Omnibus F4 and Raspberry Pi (10/30/19)
• Two way communication between vehicle and the server (11/5/19)
• User authentication (11/16/19)
• Mark drone location on map (11/24)
• Migrate to AWS RDS
• Communicate with multiple drones at the same time
• Distinguish different users. Only send vehicle updates to authorized users.
• Develop an improved UI for telemetry data.
• Use a more robust background tasks solution.

Authors

• Lyuyang Hu - Software architecture design, prototyping, project management, backend development
• Mina Bedwany - Front-end, back-end development

Advisor

• Professor Peter Burke

License

GNU General Public License v3.0