Decentralized Swarm Aware Internet of Things (IOT) protocol for real time crop quality prediction by Intelligent Autonomous Mobile Robots (IAMRs)

The increasing demand for sustainable and high-efficiency agricultural practices has accelerated the adoption of Intelligent Autonomous Mobile Robots (IAMRs) for real-time field monitoring and crop quality assessment. However, the practical deployment of such systems is constrained by limited onboard computational resources and the inefficiencies of conventional centralized Internet of Things (IoT) communication architectures. This paper presents an integrated agricultural monitoring framework that combines a resource-efficient computer vision model with a dedicated Swarm Aware Internet of Things (SA-IoT) protocol for real-time crop quality prediction using IAMRs. Unlike traditional IoT systems that rely on centralized servers for data processing and decision-making, the proposed SA-IoT protocol adopts a fully decentralized, peer-to-peer Internet of Robots (IoR) architecture, where each mobile robot operates simultaneously as a data collector, processor, and decision-making agent. Crop health is evaluated locally using a lightweight computer vision approach based on HSV color space analysis, enabling accurate plant health assessment on low-power edge hardware without the need for computationally expensive deep learning models. Extensive simulations and field experiments demonstrate that the proposed system achieves near real-time crop health mapping, robust fault tolerance through autonomous task reallocation, and significant reductions in coverage time and energy consumption compared to single-robot and centralized IoT baselines. Results show up to a 4.75× improvement in field coverage efficiency using a five-robot swarm, while maintaining high reliability and scalability. The proposed framework validates the feasibility of decentralized swarm-based agricultural monitoring and provides a practical, energy-efficient solution for precision agriculture applications.