A Review of UAV-Based Artificial Intelligence Applications in Agriculture

Integrating unmanned aerial vehicles (UAVs) with artificial intelligence (AI) is transforming modern agriculture through data-driven, site-specific management. This review syn-thesizes findings from 48 recent studies, covering three main applications: crop growth monitoring and yield estimation, pest and disease detection, and weed identification and management. UAV sensing technologies—including RGB, multispectral, hy-perspectral, and thermal sensors—are examined alongside AI-based analytical methods ranging from machine learning to deep learning architectures. Key advances include multi-sensor data fusion, temporal analysis, and strategies to mitigate data scarcity. This work provides a comparative synthesis of approaches across domains, highlighting both algorithmic developments and operational constraints. However, challenges remain, such as limited annotated datasets, weak model generalization across environments, and high deployment costs. Future priorities include multi-source data fusion, Edge AI for real-time onboard processing, and agricultural digital twins. Progress in these areas will be pivotal to building productive, resilient, and sustainable farming systems.