Integrating Precision Agriculture through Automated Nutrient Analysis and Artificial Intelligence Crop Decision Modeling

Overview

duPont Manual High School To maximize crop production many farms utilize fertilizers that supplement the soil with nutrients, such as nitrogen, potassium, and phosphorus. However, many farms over apply the necessary amount of fertilizer in order to ensure high yields. To exacerbate the issue, fertilizer costs are now exponentially increasing, with sales exceeding $18 billion dollars annually just within the United States. Lamentably, the financial cost of applying extra fertilizer in a field is substantially lower than the potential yield loss as a result of an under-application of fertilizer. However, this over application of fertilizers is mainly due to the absence of a feasible method to quantify nutrient concentration in real-time. To combat this problem I created a two pronged solution. Firstly, I created a feasible and automated method of rapid and efficient nutrient quantification for the main nutrients. Secondly, I created a machine learning and artificial intelligence algorithmic program capable of predicting crop yields utilizing the nutrient values quantified from the former automated process, as well as an input of other variables, for rural farmers that have limited access to technology, allowing users to make timely and informed day-to-day decisions. By creating predictive crop models, small scale farmers will be able to make decisions for the amount of fertilizer necessary based on the outputs created by the algorithms, thus mitigating the issues created by the over and under application of fertilizers. This two pronged solution allows for a more efficient method of fertilizer application with far less detriments on the environment.

Competition history

• JSHS 2020 Category not listed

Resources

• National JSHS abstract book ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.