Artificial Intelligence in Agriculture: Emerging Applications, Challenges and Future Directions
TL;DR
This review synthesizes recent advances in AI driven agriculture, highlighting the evolution of intelligent farming, key methodologies, applications, and future directions and emphasizes the growing importance of multimodal AI, explainable models, edge intelligence, and autonomous systems in building sustainable, climate resilient, and resource efficient farming practices that can strengthen global food security.
Abstract
Artificial Intelligence (AI) is rapidly becoming a transformative force in agriculture, enabling smarter, data driven decisions across the entire production cycle. By combining machine learning, deep learning, computer vision, remote sensing, IoT, robotics, and cloud computing, AI has accelerated precision farming practices. These advances have improved crop monitoring, disease diagnosis, yield prediction, irrigation management, nutrient optimization, and even autonomous farming systems (Wolfert et al., 2017; Liakos et al., 2018; Kamilaris & Prenafeta Boldú, 2018; Benos et al., 2021). Recent breakthroughs in transformer architectures, multimodal learning, explainable AI (XAI), digital twins, and foundation models have strengthened AI’s ability to integrate diverse datasets from satellites, UAVs, weather stations, sensors, and farm records for real time decision support (Basso & Antle, 2020; Shahhosseini et al., 2021; Dainelli et al., 2022). Compared with traditional statistical methods, AI consistently delivers superior results in yield forecasting, disease detection, irrigation planning, weed identification, and resource optimization by modelling complex interactions among climate, soil, and management factors (Khaki et al., 2020; Talaviya et al., 2020; Feng et al., 2021). However, challenges remain. Limited access to high quality datasets, difficulties in transferring models across regions, computational demands, lack of transparency, data governance concerns, and socioeconomic barriers continue to slow adoption (Benos et al., 2021; Guidotti et al., 2018; UNESCO, 2021). This review synthesizes recent advances in AI driven agriculture, highlighting the evolution of intelligent farming, key methodologies, applications, and future directions. It emphasizes the growing importance of multimodal AI, explainable models, edge intelligence, and autonomous systems in building sustainable, climate resilient, and resource efficient farming practices that can strengthen global food security.