Differentiation of Mamdani and Sugeno Fuzzy Inference System in developing Automatic Plant Watering Systems for Domestic Use

Abstract

Water management systems are an essential requirement for maintaining plants. One of the major problems encountered in growing plants in houses, offices or buildings with in-house kitchen gardens is irregular irrigation patterns. This leads to over irrigation of pots or elongated durations of dry soil, both conditions ill for plant growth. This problem is further aggravated when manual irrigation ignores the environmental conditions and plant specific requirements. Utilizing Fuzzy logic for developing Automatic Plant watering systems provide flexibility in manipulating input parameters such as Temperature, Humidity, Soil Moisture and Plant Growth to determine optimum flow rate of the irrigation system. In this study, multiple fuzzy systems are developed for different environments and parameters. MATLAB is used for designing the fuzzy logic controllers using both Mamdani and Sugeno Models. The requirement for the system is to adjust the flow rate in accordance with the environmental conditions and plant requirements. The paper draws comparison between Mamdani and Sugeno methods on the basis of their performance characteristics for different environments. It also provides development of an effective controller for a watering system for household plantations.