Farming is becoming more and more dependent on technology. It would be extremely beneficial to see the status of all the machinery in an easy-to-use dashboard.
What it does
We made a dashboard that displays live critical information about various farm machinery. It also has a built-in anomaly detection system, meaning that it alerts you when your machinery's sensors are displaying something out of the ordinary. This allows you to detect potential errors early, saving millions of dollars in lost time and repairs.
How we built it
We built it using AWS, Scikit-learn, and Express.js. The model was trained locally, and wrapped in a lambda function. The webpage is loaded from S3 also using lambda functions.
Challenges we ran into
None of us had ever used AWS before, so it was a challenge to get all the services up and running.
Accomplishments that we're proud of
All the AWS stuff! We learned that AWS is powerful, but also frustrating without proper help.
What we learned
We learned that sometimes it is better to use pretrained models as opposed to creating our own.
What's next for Tractor Dash
At the start we wanted to incorporate aggregate data and environmental impact into our app. Also, we would like to eventually incorporate this dashboard into the Fuse Smart Farming app.