A lot of drivers are frustrated about finding a good parking spot, especially in downtown areas, so our team decided to create an application that would help them find the closest parking spot given their destination. This way, they spend less time wandering around for a parking spot.

What it does

This application acts as a navigator that will direct the user to the nearest parking spot to their destination and display a walking route from that parking area to the destination. Keep in mind this application only takes into account London parking data.

How we built it

We used Flask as the framework to build our web application, Google Maps API to display routes for the user, SKLearn to classify groups of parking spaces, FireStore to host our database, and Google Cloud to host our Flask web application.

Challenges we ran into

None of us have any experience with Flask and Firebase products so as a result, we struggled to deploy the web application to firebase for hosting purposes. In addition, it required a significant amount of effort to set up the database because the parking space data needed to be cleaned, processed with machine learning, and then stored for future use by the Flask application.

Accomplishments that we proud of

Finished this project in a day and a half

What I learned

How to choose the right machine learning method to solve real-life problems even with simple data like coordinates. We learned how to use APIs in conjunction with each other to extract and display the data we want. We also learned more about the end-to-end process of web development from setting up the database, developing the application, and designing the front-end.

What's next for NavPark

A viable feature for the future would be to provide multiple parking spaces for users so if one parking space is full when they arrive, the user can request for NavPark to redirect them to the next closest parking spot. Another robust feature would be for NavPark to determine if certain areas for parking, especially parking meters, will be all taken at a certain time and whether to ignore these areas from NavPark's considered parking spaces.

Share this project: