Low-power IoT Networks for Autonomous Farm Robotics and Sensor Fusion: A Comprehensive Review

Abstract

The integration of low-power internet of things (IoT) networks, autonomous farm robotics, and multimodal sensor fusion is rapidly transforming modern agriculture into a data-driven, intelligent, and sustainable ecosystem. Low-power wide-area networks (LPWANs) such as LoRaWAN, NB-IoT, and Sigfox play a critical role by enabling large-scale, long-range, and energy-efficient communication among distributed sensor nodes deployed for soil, crop, environmental, and livestock monitoring, even in remote rural regions. At the same time, autonomous agricultural robots, including unmanned aerial vehicles (UAVs), unmanned ground vehicles (UGVs), and multi-arm robotic manipulators, are increasingly used to automate labor-intensive tasks such as precision seeding, targeted spraying, mechanical weeding, harvesting, crop scouting, and yield estimation, thereby improving productivity and reducing resource wastage. Multimodal sensor fusion techniques that combine data from ground-based sensors, RGB and multispectral vision systems, LiDAR, hyperspectral imaging, and localized weather stations enable real-time situational awareness, robust perception, and accurate decision-making under dynamic field conditions. This review synthesizes findings from more than fifty peer-reviewed studies published between 2020 and 2025, focusing on system architectures, communication protocols, edge- and cloud-based data processing, AI-driven analytics, and intelligent agricultural robotics. Key challenges such as energy optimization, intermittent rural connectivity, latency constraints, interoperability across heterogeneous devices, multi-robot coordination, and scalability in large-scale deployments are critically examined. Furthermore, emerging research directions, including 6G-enabled robotic swarms, digital twin-based farm modeling, nano-IoT sensing, soft robotics, and intelligent edge computing, are explored as promising pathways for next-generation smart farming systems. Overall, this work presents a unified conceptual framework to guide future research and practical deployment of low-power IoT-enabled autonomous agriculture.