Advances in intelligent and autonomous greenhouse systems: A comprehensive review of internet of things, artificial intelligence, and robotics integration

Abstract

• An analysis of the current research status of the latest advanced technology has been conducted. • Advanced technology systems hold promise for detecting combined drought stress and the behaviours of plants. • It is necessary to explore the critical pathways of how the advanced technology is used in the agricultural control environment. • There is a need to develop a simple yet efficient system that the farmers can use in their fields rather than using a complex and difficult system that needs experts. Arid and hyper-arid regions are experiencing continual losses of arable land due to environmental degradation, climate change, rapid population growth, and industrial expansion. Greenhouses equipped with advanced monitoring and control technologies offer a promising strategy to sustain and enhance agricultural productivity in these fragile environments. Recent studies reveal a sharp increase in the development of greenhouse monitoring systems; however, their practical application remains limited due to integration complexities, data scarcity, and computational constraints. This review synthesizes over 100 studies published since 2021, focusing on the application of technologies such as the Internet of Things (IoT), machine learning (ML), deep learning (DL), and robotics for yield prediction, disease detection, and plant growth monitoring. All reported datasets, artificial intelligence (AI) models, and their corresponding performance metrics were critically examined. Overall, this review identifies, categorizes, and evaluates current and emerging trends in intelligent greenhouse monitoring and control systems to advance research and innovation in this field. Technologies such as IoT, ML/DL, and robotics are recognized as key innovations that advance unmanned greenhouse management while improving energy and water efficiency. In addition to analyzing cutting-edge technologies in the greenhouse sector, this review also addresses persistent challenges—including multimodal sensor fusion, model adaptability, limited data availability, and computational constraints—along with future trends and research opportunities. This study highlights the significant potential of IoT, ML/DL, and robotics to accelerate the global development of data-driven, intelligent greenhouse applications.