Fuzzy Inference System and IoT Based Smart Irrigation for Smallholder Agriculture in Rural Bangladesh

Agriculture constitutes one of the cornerstones of Bangladesh's social, economic, and political well-being. It is the important sector in terms of water consumption and waste, owing to inefficient irrigation techniques. In fact, poor mitigation strategies in cases of over- or under-irrigation have resulted in a drop in production rates. The purpose of this paper is to propose a Fuzzy Inference System (FIS) based smart irrigation mechanism for smallholder agriculture in rural Bangladesh that works to reduce irrigation frequency while increasing production rate. The system consists of a Mamdani fuzzy inference controller that gathers data from various sensors, including those measuring soil moisture, humidity, light intensity, and temperature. It uses a set of rules in a fuzzy inference system to control the flow of water from the water pump and activate irrigation at appropriate intervals. The system is simple to use and financially viable. It remains advantageous even when considering vast agricultural fields, as it effectively reduces water and energy consumption.