AI-Driven Autonomous Robot for Crop Monitoring and Precision Farming

Agriculture plays an important role in food production and economic growth. However, crop diseases, improper soil moisture, and changing weather conditions can negatively affect crop health and reduce agricultural productivity. In many farming practices, farmers monitor crops through manual inspection, which is time-consuming and can delay disease detection. To address these challenges, this research proposes a low-cost, ground-based agricultural robot that combines deep learning with environmental monitoring to support precision farming. The proposed system uses an existing YOLOv8-based model for crop disease detection, trained and evaluated using the PlantVillage dataset. The model achieved a reported test accuracy of 99.76% on the current test set. However, this result is based on testing on the dataset, and further evaluation is needed to assess performance under real-world field conditions. A four-wheel mobile robot equipped with a camera is planned to capture plant images at close range, while a laptop will process the images and predict possible crop diseases. An ESP32-based sensor module will also be integrated to monitor soil moisture, air temperature, relative humidity, rainfall, and water-tank levels. The collected sensor readings and disease detection results will be displayed on a dashboard, which will provide basic recommendations using predefined rules. At present, the machine-learning model has been developed and tested using plant images. The construction of the robot, integration of sensors and dashboard, and testing in real agricultural environments are planned as the next stages of development. The proposed system aims to offer an affordable and modular solution for crop monitoring, early disease identification, and agricultural decision-making.