Agrisentinel Rover: An IoT-Integrated Ml System for Crop Detection and Plant Disease Analysis

Agriculture plays a fundamental role in global food production; however, plant diseases, climate variability, and delayed monitoring significantly reduce crop yield and quality. Traditional manual inspection methods are labor-intensive, time-consuming, and prone to human error, particularly in large-scale agricultural environments. This paper proposes the AgriSentinel Rover, an autonomous IoT-integrated robotic system designed for real-time crop detection and plant disease analysis using deep learning techniques. The system integrates environmental sensors, high-resolution image acquisition, Convolutional Neural Network (CNN)-based classification, and cloud-based monitoring to enable predictive and precision agriculture. The rover autonomously navigates agricultural fields, captures crop leaf images, collects environmental parameters such as soil moisture, temperature, humidity, and light intensity, and processes the data using a trained MobileNetV2-based model. The processed information is transmitted to a cloud dashboard, providing real-time alerts and analytical insights to farmers. Experimental evaluation demonstrates high disease classification accuracy and efficient environmental monitoring, thereby improving early detection and reducing crop loss.