Multiple Sensor Data Fusion Based Automatic Plant Pot Watering Employing Mapping by Linear Regression

Abstract

An automaton for calculating the quantity of water to be released to the plants periodically is calculated by a linear regression model with two input variables and one output variable. The two input variables are the reading of two sensors: the soil water moisture content sensor and the photoresistor sensor. The information from these sensors is collected and the manually released quantities of water are fed as training data set for a Linear Regression model. The trained model is used to actuate the water releasing actuator.