-
-
testing optimization code
-
Graph Visualization
Inspiration
Our team saw NextEras challenge as one of the more interesting challenges out of the bunch. We are pretty passionate about drone technology and though it would be quite the challenge to learn about route optimization.
What it does
The project optimizes flight routes for cost efficiency so that it can complete all inspection missions while respecting battery constraints. It also creates visualizations to show each individual mission path and the points it's traveling to.
How we built it
We built it using Python as the primary language while simultaneous using docker to set up an environment with all the necessary dependencies.
Challenges we ran into
We had initially planned to do the project in Java but found that it was better to do the optimization in Python, due to the available libraries, so we had to convert much of our initial code to Python.
Accomplishments that we're proud of
We managed to get the visualizations working, and optimize the algorithm to look at multiple potential routes and select the most efficient one, within a limited amount of paths. It selects how many paths to look at, based on the amount of data inputed.
What we learned
We learned to use various APIs such as Plotly, Shapely, and Geopandas. We learned how to utilize open-source platforms like Docker, as a part of our data pipeline. We learned how to utilize algorithm optimization libraries like Google OCR, and adjust them to our needs. We also learned how to use AI tools, like Gemini, to make our workflow more efficient.
What's next for Drone Optimization Challenge
In the future, we'd hope to further optimize it, by running it on multiple CPUs, so that we can compare more potential paths, hopefully finding a more efficient path, without adding additional time to our runtime. We'd love to add more constraints to our algorithm, like weather, so that our potential paths can better accommodate real-life issues.
Log in or sign up for Devpost to join the conversation.