ISSN 2063-5346
For urgent queries please contact : +918130348310

Machine Learning based IoT application to Improve the Quality and precision in Agricultural System

Main Article Content

Venkata Kanaka Srivani Maddala , Dr. Shantanu Shahi, Dr. Yusuf Perwej , H G Govardhana Reddy , Mr. Utkarsh Arun Avalekar, Mirzanur Rahman
» doi: 10.31838/ecb/2023.12.si6.157


The backbone of the Indian economy is agriculture, and since crop productivity is heavily reliant on natural factors, nature plays a key role in this system. Although it is difficult to alter the natural environment to suit agricultural needs, crops may be chosen based on the current state of the environment. The majority of farmers in India operate on a small or marginal size, hence it is unlikely that they would consult professionals or any other informational source to choose the best crop or crops. We might conclude that there is a disconnect between farmers and current research. The data that is currently available (reports) tends to be static and cropand duration-specific. There is thus no assurance that it will function well for every farm in the present environment or in every scenario. Here are several hardware and software solutions to the problems mentioned above. The "Internet of Things" is a current trend, and it is suggested here to use it to continually monitor the agricultural field using the "Agriculture Monitoring Model" (AMM), which comprehends and records the soil-environmental conditions. AMM is a lightweight, affordable model. The continually collected data was further analysed for decision-making, acting as a software model for an advising system. A current development is machine learning, which uses both supervised and unsupervised methods. The suggested system was created using a supervised approach, which takes into account previous data while building the model and making judgements for the present situation. The model is taught to be more accurate the more balanced the input data, and the more correctly it functions for both current and future inputs. The phrase "present inputs" refers to the already accessible natural characteristics, of which only a small number may be purposefully controlled. The AAS (Agriculture Advisory System) will make a determination on the degree of crop suitability and the necessary fertiliser (regular/micro-dose). Farmers incur costs while micro-dosing seeds and fertiliser since there aren't enough devices available when they're needed. Here, a low-cost IoTbased AgriRobot is presented to automate the sowing and micro-dose fertiliser mechanism, reducing the issue of labour reliance

Article Details