MAXIMIZING AGRICULTURAL WATER EFFICIENCY: INTEGRATING IOT AND SUPERVISED LEARNING FOR SMART IRRIGATION OPTIMIZATION

Abstract

Optimum utilization of clean water around the globe is essential in order to avoid scarcity. In agriculture, due to the lack of intelligent irrigation systems, consumes more amount of fresh water. Smart irrigation using IoT technologies can solve the problem by achieving effective utilization of water. By examining ground parameters such soil temperature, air moisture, soil moisture, humidity, and weather data (precipitation) from the website, this research project forecasts the irrigation schedule. When designing intelligent irrigation, soil moisture is a key consideration. It is suggested that a hybrid machine learning algorithm be used to estimate the soil moisture for the next days using field, environmental, and weather data in order to accomplish smart irrigation. The field data are gathered by sensors and are transmitted via wifi to the server and the web-based interface is developed to visualize the current field data, weather data, and schedule of the next irrigation. The system is fully autonomous which starts and stops the irrigation based on the result of the algorithm. This work depicts the architecture of the system and describes the information processing of the results for a month. The accuracy of the propsed algorithm is good and has a minimum error rate of predicted soil moisture.