Design of Crop Recommender System using Machine Learning and IoT

Abstract

Agriculture is one of the key drivers of Indian economy. The primary problem now confronting Indian farmers is that farmers don't choose the right crop based on their land requirements. A significant decline in production is seen as a result. Precision agriculture will provide the farmers with a solution to this problem. To suggest the optimal crop to farmers based on site-specific criteria, precision agriculture uses research data on soil types, features, and crop yields. With the help of an intelligent system, this study aims to help Indian farmers increase crop productivity by selecting the right type of soil. The proposed prototype considers soil characteristics, such as pH value, soil temperature, and soil moisture, as well as environmental factors, such as humidity, as inputs to the machine learning algorithm for decision-making. The output is integrated with the web program known as proteus. The entire prototype is designed using STM32 ARM Processor and simulated using proteus, and the same is implemented using the Nucleo board by integrating the humidity, pH, and temperature sensors for collecting the input data. The result of the prototype is also displayed in the Blynk app as well as the LCD display, where the system recommends the appropriate crop.