What it does
My tool allows users to simply navigate through buildings given just an annotated floor-plan. It does this by utilizing basic graph theory, solving algorithms like A*STAR, and by creating a weighted map of every path of travel in the building.
How we built it
I annotated the first two floors of Swem, added data on how each room connected to the other, then connected the two floors together via stairs or an elevator. This is extensible to any building on campus - it just requires some initial work of annotating the floor plans.
Challenges we ran into
One big challenge was how to measure what the "best move" was. We utilized the coordinates and simply calculated distance traveled from room to room - the rooms were always simple enough for this heuristic to succeed.
Accomplishments that we're proud of
Implementing an unweighted graph and annotating the floor plan took up a lot of my time!
What we learned
How A*STAR works, basic graph theory, indoor navigation & wayfinding, etc.
What's next for W&M NavigAider
Next, I plan on allowing a more simple building configuration file as an input. Additionally I really want to implement an actual visualization - currently the map is shown, and I have the variables waiting in the webpage to just render points between them, but I just did not get it done in time. But, the data is already there for me to do it.
Built With
- astar
- graph
- rust
Log in or sign up for Devpost to join the conversation.