Integration of Artificial Intelligence with Real-Time IoT Based Distributed Edge-Cloud Frameworks: Applications, Challenges, and Solutions

Abstract

This systematic literature review investigates the convergence of artificial intelligence (AI) and real-time Internet of things (IoT) systems within a distributed computing environment. Covering research from 2021 to 2025, the study highlights how AI-enabled real-time IoT systems enhance intelligent, latency-sensitive applications across various domains, including healthcare, smart cities, industrial automation, agriculture, and public safety. The review focuses on the use of the main technologies such as federated learning, reinforcement learning, tiny machine learning, transfer learning, anomaly detection, and self-supervised learning, to overcome challenges related to latency, resource constraints, data heterogeneity, interoperability, scalability, and security. This article also highlights the main challenges and solutions to those challenges with the different methods and techniques. By presenting taxonomy tables, use case analyses, and cross-domain comparisons, the paper outlines different solutions that enable dependable and scalable AI-IoT integration. Furthermore, it explores the emerging role of future technologies, sixth-generation, digital twins, and the metaverse in transforming edge AI infrastructures for secure, efficient, and real-time decision-making. The findings provide a roadmap for researchers and practitioners to design robust, intelligent, and context-aware distributed systems for next-generation applications