Autonomously landing a drone on a visual target. This project was built to prototype an algorithm for the Mohammad Bin Zayed International Robotics Competition 2017.
The project makes use of the dronekit-python library to connect to a Pixhawk based drone. Dronekit-sitl is used to validate the algorithm in a simulated environment. The target tracking is performed using the Viola Jones approach for Rapid Object Detection using HAAR-cascades.
Create a directory called Vids/
in Logs/
- Run the code with
python main.py
to run on simulation. - Run the code with
python main.py --connect <connection_string>
to run on the drone.
- Daniel Nugent's work on precision landing - https://github.com/djnugent/Precland.
- Viola and Jones' paper - Rapid Object Detection using a Boosted Cascade of Simple Features.
Disclaimer - This repository is no longer maintained.