Artificial Intelligence and Machine Learning in Precision Agriculture: Applications, Challenges, and Future Directions
S. Vishnupriya *
Department of Genetics and Plant Breeding, Uttar Banga Krishi Vishwavidhyalaya, Pundibaro, Cooch Behar, West Bengal - 736 165, India.
Rohan S Kavalagi
University of Agricultural Sciences, Raichur, India.
Amritendu Misra
Department of Genetics and Plant Breeding, Suresh Gyan Vihar University, Jaipur, Rajasthan, India.
Raosaheb Bapurao Shid
Dr. D Y Patil College of Agriculture Business Management, Akurdi, Pune, Maharashtra, India.
Alok Kumar Singh
Department of Plant Pathology, Faculty of Agriculture, Udai Pratap College, Varanasi- 221002 (Uttar Pradesh), India.
Varsha Kanojia
Processing and Food Engineering, Punjab Agricultural University, Ludhiana, India.
S. M. Bharthisha
Department of Agronomy, College of Agriculture, UAS, Dharwad, India.
S. M. Kishore
Department of Entomology, College of Agriculture, KSNUAHS, Shivamogga, India.
*Author to whom correspondence should be addressed.
Abstract
Precision agriculture has shifted from a sensor-driven discipline into a data-driven one, with artificial intelligence and machine learning now embedded in crop monitoring, disease and weed management, irrigation scheduling, soil assessment, robotics, and farm decision support. This review synthesises the state of artificial intelligence and machine learning applications across the precision agriculture value chain, drawing on peer-reviewed literature published between January 2018 to March 2026, supplemented by a small number of earlier foundational studies. The review traces the evolution from conventional algorithms such as random forests and support vector machines toward deep convolutional and recurrent architectures, and more recently toward transformer-based models, vision-language systems, and large language models adapted for agronomic reasoning. Five themes receive particular attention: crop monitoring and yield forecasting; detection and management of diseases, pests, and weeds; soil, water, and nutrient management; remote sensing and robotic field operations; and the data infrastructure, including the Internet of Things, big data analytics, and federated learning, that underpins these applications. The review also examines the economic and socio-political dimensions of adoption, particularly the disparities between well-resourced and smallholder farming systems, alongside the ethical and governance questions raised by farm data ownership and algorithmic transparency. The discussion closes with an assessment of emerging directions, including explainable artificial intelligence, privacy-preserving learning, and generative model integration, and concludes by identifying methodological gaps that constrain translation of laboratory performance into field-scale reliability. The review is intended for agricultural scientists, data scientists, and policymakers seeking a structured, critically evaluated synthesis of where artificial intelligence currently stands in precision agriculture and where it is heading.
Keywords: Artificial intelligence, machine learning, precision agriculture, deep learning, smart farming, digital agriculture