AI Integration into an Autonomous Irrigation System Optimized for Water Conservation

Overview

In this day and age where data has become our most valuable asset, it is appropriate for us to apply our ever-evolving data analysis tools and artificial intelligence algorithms to reduce the vast amounts of water wasted through agriculture. Currently, over 40% of the world’s total freshwater consumption is wasted through agriculture and although on a small scale, this project attempts to demonstrate how the implementation of artificial intelligence in a micro-irrigation system that utilizes HSV color input as a direct reflection of a plant’s health will result in the growth of healthy plants while minimizing water wastage. A preliminary set of plants is grown using a basic micro-irrigation system with no AI functionality, giving each plant different volumes of water and measuring each plant’s HSV color at the end of the growth period. User input is also required for the program to interpret the color input. This data is used to calculate a minimum healthy color threshold based on user preference and is then fed into an AI program that utilizes dynamic linear regression to analyze trends in the data and ensure that the least amount of water is used to maintain healthy plants. The regression algorithm uses watering volume in mL as the domain and HSV color as the range to calculate the slope of the linear equation, which is interpreted to be the “average color gainer per mL of water given.” This value is used to specifically adjust the watering volume when a plant is significantly above or below the healthy threshold to keep the plant’s color as close to the threshold as possible, thus keeping water wastage to a minimum. As the AI program progresses and the data pool grows, the regression algorithm is capable of continually collecting and analyzing new data points and make more accurate predictions as the data set grows. At the conclusion of the experiment, the p-value was 0.00013. The value was less than 0.05, which rejected the null hypothesis and it was concluded that there is a significant difference between the effectiveness of watering using a micro-irrigation system with artificial intelligence implementation and human watering.

Competition history

• AJAS 2020 Category not listed

Related projects

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

Browse more like this

• All Environmental Engineering projects
• AJAS Environmental Engineering projects
• Every AJAS 2020 project