OBJECTIVE
The goal of the project was to make the drone capable of following a moving object. We had the freedom of choosing the object. The drone had to keep track of the object while a judge kept moving it around.
The team took the decision of using a blue cardboard since a solid color would make it easier for the drone to get closer. This part of the algorithm made use of OpenCV to detect the color, and area of the figure, the area allowed us to know the approximate distance to the target. Finally, to send the movement instructions we made use of ROS.
In the center of the cardboard a QR code was placed to make the algorithm more robust against possible similar objects in the environment. ZBar was used to make sure that the QR code was correct and keep track of it with the help of OpenCV.
After the successful testing (shown in the Gallery Section), a cardboard cube was going to be manufactured to avoid losing the target in case that the judge decided to rotate the object. However, due to the cancellation of the tournament as a consequence of the current pandemic, and to avoid unnecessary health risks, the project was cancelled.
TECHNICAL DETAILS
- Python
- Ubuntu OS
- Adaptability
- Problem Solving
- OpenCV
- Computer Vision
- ROS
- Self-learning
- Time Management
- Critical Thinking
- Teamwork
- Openness to Criticism
- Oral Communication
This project was made using:
GALLERY
Drone Demo Video
The full demo video is available here
Drone demo key moments
Drone Initial Position
First Detection of the Target
Drone Enters QR Tracking Stage when Close Enough to Target
