Development of Artificial Intelligence Algorithm for Smart Irrigation Using Internet of Things (IoT)

Abstract

Artificial Intelligence (AI) is the most recent agricultural technology. In agriculture, water is used to irrigate the plants so that they can flourish. Due to the scarcity of water in most parts of the world, the watering process is one of the most significant and crucial procedures. To address this issue, a smart irrigation system based on the Internet of Things (IoT) was developed utilizing AI technology, an Arduino Uno microcontroller, and sensors. The objective of this study is to develop a modified ANFIS (Adaptive Neuro-Fuzzy Inference System) AI algorithm for improved automated irrigation system decision control and to reduce the computing complexity of ANFIS architectural layers. It also seeks to develop an integrated system for monitoring and managing irrigation to increase agricultural output using MANFIS and the Internet of Things. Lastly, is to determine the difference of algorithmic complexity between the conventional ANFIS and Modified ANFIS. As a result, architectural layers were reduced into 3 layers, INPUT, PROCESS and OUTPUT. It also waters the plant automatically and sends signals regarding smart irrigation system information such as the tank's water level, the plant's soil moisture content, and the trigging factor, or the quantity of water to be released which enables the farmers to monitor and manage its irrigation system using the IoT. The simulations were carried out using MATLAB software, a fuzzy logic controller is used to control the whole system by providing its input, rules and output. To determine the computational complexity of each method, the ANFIS and modified ANFIS were examined with 100 % success rate.