A systematic literature review of artificial intelligence and internet of things enabled crop farming

Résumé

Abstract The merging of Artificial Intelligence (AI) and the Internet of Things (IoT) has generated significant scholarly interest as a means of improving productivity, sustainability, and resilience in modern crop farming. This systematic literature review (SLR) examines recent studies published between 2022 and 2025 that focus on the development and evaluation of Artificial Intelligence of Things (AIoT) technologies in agricultural systems. Adhering to the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), the results are synthesized in this review in response to four research questions focusing on AI methodologies, IoT architectures, data acquisition and management strategies, and research challenges that remain outstanding. The findings show that machine learning models and deep learning models have been widely used in various applications such as crop recommendation, predicting yield, optimisation of irrigation systems, and disease and pest detection. Convolutional neural networks and approaches based on ensemble learning dominate vision based tasks, and light-weight machine learning approaches and federated learning techniques are also subject to increasing research to enable privacy preserving and edge-based deployment. Architecturally, one can observe a certain shift in picking up from cloud-based only solutions to hybrid edge-fog-cloud frameworks to enable low-latency and better scalability. Effective data acquisition is dependent on the use of multimodal sensing and data fusion with the support of edge level preprocessing and cloud based analytics, it is also seen how emerging technologies like blockchain and InterPlanetary File System (IPFS) can be used for decentralised data acquisition to improve data security and traceability. Despite these improvements, however, the review points to ongoing barriers such as the bias in datasets, lack of field-scale validation, computational and energy constraints, interoperability problems, and gaps between experimental prototypes and real-world deployment. By bringing together recent evidence and recent open research directions, the current review provides an organized and updated reference for informing the design and development of scalable, secure, and deployment-aware AI-based IoT-based crop farming systems. Clinical trial registration Not applicable.

Citer ce document

Accès au document

Texte intégral en lecture en ligne, réservé aux abonnés SPHAERO et aux membres de l'institution. Se connecter

Auteur(s)

Statistiques

Consultations : 1

Téléchargements : 0