Inspiration
What it does
Each drone acts as an individual entity, which tracks other nearby drones and adjusts course to avoid collisions.
How we built it
The solution was initially implemented as an embedded system using QNX. However, due to software difficulties, we were forced to pivot to a purely simulated approach. This consisted of a GUI and simulator written in python, which communicated with multiple drone models, each following a policy to prevent collisions while minimizing flight time.
Challenges we ran into
We encountered significant difficulties while using QNX, eventually forcing us to pivot. Additionally, integrating the Python Tkinter GUI with the individual drone models written in C++ proved to be challenging.
Accomplishments that we're proud of
We are very proud of the fact that we created a novel solution to a problem that we have not yet seen discussed in industry. Seeing as this was our first hackathon, we are all very pleased with what our team was able to accomplish.
What we learned
We learned that knowing your tools is one of the most important factors in delivering a working product. We also learned how to cooperate in a software team under high time pressure. Individuals on our team also broadened their know knowledge of specific tools such as Git and Docker, and libraries such as TKinter.
What's next for Real-Time Drone Collision Avoidance
We hope to continue the project on our time, because we believe in the importance of the software, and are interested in honing our skills further. We want to attempt an embedded solution again, and scale to multiple embedded systems to fully realize the communication software we developed.
Log in or sign up for Devpost to join the conversation.