Plowee | Devpost

Our logo.
The raspberry pi in development.
The overall whiteboard cooking area. We had a lot of ideas...
The application used to track the ice-related dangers.
The management app that community leaders can use to organize the snow plow movements.
The boys hard at work.

Inspiration:

Ice poses a significant hazard in daily life, especially on roads and sidewalks, leading to accidents and delays. By providing real-time snow plow tracking and ice condition reports, Plowee enhances safety and helps communities navigate winter conditions more effectively. Our project poses to present a set of mobile applications dedicated to giving our communities management over a massive danger: ice.

What it does:

Our set of applications, plus a hardware integration directly for the snow plows, allows users be notified and to notify snow plow management that roads/sidewalks in their current location need to be cleaned up. Additionally, users are able to enter a destination they would like to reach, and we did a lot of debugging and acquiring of technical expertise related to building the signed APK and linking of the data servers. Overall, an incredibly tough project to integrate all together: and still much more work to be done.

How we built it:

Using a lot of tech stacks, including Flutter/Dart, OpenAI, SQL, Google Maps APIs, Raspberry PI, and many more.

Challenges we ran into:

We encountered massive challenges when working with older hardware available to simulate the snow plow vehicle movement. A hardware board that came out literally 10 years ago, connecting the Raspberry Pi to the internet for wireless connection and transmission of location-based data. The board, essential, was running into many issues and was could only connect low-broadband internet at first. However, determined to figure out a way we could recycle these old boards into a modern program, we debugged and used a few external library commands that helped us overcome this challenge. Additionally, the desire to develop in React Native, while initially fun and a learning experience, proved too difficult to implement across all of our computers for a multi-platform software. However, quickly switching to Flutter, we directly created a working base app in 30 minutes. This did come at a cost, as the runtime for running the React Native app took over 7 hours and we essentially started from scratch.

Accomplishments that we're proud of:

Overall, we successfully implemented the overall OpenAI feature for detecting road intersections and potential danger areas, as well as linkage of Supabase and SQL server querying for the first time. Of course, using Raspberry Pi to track the accurate position of the plower using WiFi, entirely disconnected from the computer, was of upmost celebration: reintegrating an otherwise obsolete and the next pollutant for communities around the globe.

What we learned:

We learned a lot about React Native, Raspberry Pi code, server-side implementation and SQL, OpenAI in the use of data analysis, and creating a full-fledged set of Flutter apps that we are proud to release in beta here for the world to enjoy.

What's next for Plowee:

There are yet many features still need to be implemented, as well as a re-physical configuring of the Raspberry Pi board for easier across-the-board use, and things like SMS sending based in the Manager app, ice route toggling, and large-scale plow object tracking are major features that will make our project, and the people it impacts, all the better off. Please leave a review and let us know what you think! We really enjoyed this opportunity and would appreciate any feedback in the comments of things we could add or focus on. Happy hacking!