Inspiration
NavCanada is transitioning to a larger picture view of canadian air traffic, which requires scheduling tools which can analyse flights for their entire trip accross canadian air space rather than in sections. This is a difficult task which would be difficult for humans alone ot accomplish, so we aim to provide a tool that will assist them in this transisiton.
What it does
Nav Canada Analyser (NCA) extracts flight paths from a given schedule and compares them to find possible conflicts and hotspots. It displays statistical insights about them, and calculates an alternative schedule free of conflicts.
How we built it
The app is powered by a python api, which performs the extraction and analysis. The insights are displayed on our React website and the modified schedule is returned as a json in the same format the origional schedule was recieved in.
Challenges we ran into
We origionally designed the conflict detection to scan in 15min intervals for efficiency but there were conflicts we missed, so we redesigned the program to first extract flight paths as lines in 4d space. Using those paths we could find the exact spacial and temporal coordinates of any conflict no matter how short the duration.
Accomplishments that we're proud of
In a weekend we managed to implement a fairly complex model of a flight path with no prior work with that kind of data. And we managed to put together a substantial list of insights for the time availible. Not to mention adding a schedule creator on top of that.
What we learned
We learned about the complicated process NavCanada must follow to keep our airspace safe and efficient, and about the sophisticated tools they employ to do so.
What's next for NAV CANADA ANALYZER
As nexts steps we'd like to improve the schedule creator function to take into consideration a larger variety of cost factors. For example we consider very basic concepts of fuel efficiency and finding short routes, but we do not incorperate weather data. We'd also like to provide more insights about times and areas with no traffic or light traffic to aid controllers in modifying the schedule, as at the moment we primarily provide insights about the conflicts themselves. Lastly we'd like to integrate Gemini's api to provide more human looking responces to clarify our insights.
Log in or sign up for Devpost to join the conversation.