Table 40: Jason Huang, Thomas Nguyen, Shelley Wang


Agriculture is largely dependent on environmental factors, and farmers can only do so much to battle against natural forces, often resulting in unmet expected crop yields. Moreover, farmland is often overwatered and fertilized, leading to fertilizer runoff and large financial losses. Not only is this uneconomical, but it also pollutes surrounding bodies of water and is vastly unsustainable. To approach these inefficiencies, we generated a multitude of data organization ideas that could also function as a prediction tool for farmers at all levels. To predict the future, we must also look at the past. FarmTrackr is user-oriented, efficient, multifaceted, and feasible.

What it does

FarmTrackr is more than a tool for farmers to archive and view personalized agricultural data. It uses environmental data to estimate the harvest rate, date, and yield. Moreover, it predicts a customized watering and fertilizing plan for maximizing crop yield while minimizing waste. By keeping a cloud-based almanac of empirical farm data, farmers can also share this in-depth information to successors and business partners. FarmTrackr sets your farm on track.

How we built it

Our team build our product using primarily web based languages (HTML, CSS, JavaScript). We used the framework Meteor.js to run our web application.

Challenges we ran into

Our team was comprised of three members with varying levels and backgrounds in programming. With the time constraint, we ran into a lot of hiccups where we had to spend time researching not only agriculture but also javascript as well. Communication was a struggle, as Jason had already been experienced in web design, and the rest of us had just begun to familiarize with it. Moreover, none of us had extensive back-end knowledge, so the entire process was a lot of learning as we went. We ended up designating roles systematically to highlight our strengths in contribution to the project. The biggest challenge of this project was to spend time learning while programming against time.

Accomplishments that we're proud of

While we all learned separate things at separate levels, we are all proud of the effort we all added into this project. Jason is proud of our overall product demo and design, Thomas is proud that he now has a better understanding and feels more comfortable with HTML and CSS, and Shelley is proud to be able to put her design major to use through graphics while applying the communication skills her computer science minor is intended to do.

What we learned

Overall, we learned that it is extremely difficult to work at varying levels under a time constraint. While those with less experience largely depend on those with more experience, those with more experience cannot rely on anyone. However, this can also promote faster learning and collaboration under pressed conditions. It also pushes everyone to contribute their maximum effort.

What's next for FarmTrackr

FarmTrackr has lots of room for expansion. While its algorithm currently makes predictions based off of environmental data, once users begin inputting their farm data, we would implement an algorithm that improves prediction accuracy based off of their personalized farming practices and input data. Moreover, we would compile input data from farmers who are near each other to improve each of their prediction accuracies, fostering a collaborative process. In the long run, FarmTrackr plans to implement a mapping system that autosizes to ratio and allows farmers to either manually map out their farm or automatically fits a best-fit map for them to base their farm off of.

Share this project: