Artificial Intelligence-Augmented Edge Computing: Architectures, Challenges, and Future Directions
Résumé
Artificial Intelligence-Augmented Edge Computing (AI-AEC) is an emerging paradigm that combines the computational intelligence of AI with the proximity, efficiency, and responsiveness of edge computing. This integration enables real-time data processing, improved privacy, and reduced network latency by bringing intelligent computation closer to data sources. As the demand for low-latency and context-aware applications grows across sectors such as healthcare, transportation, smart cities, and industry, AI-AEC presents a transformative approach to addressing these requirements. This article provides a comprehensive overview of the core architectures that support AI-AEC, including hierarchical, collaborative, and decentralized models. It examines key enabling technologies such as federated learning, lightweight AI models, edge hardware accelerators, and high-speed connectivity frameworks like 5G and beyond. The paper also identifies and analyzes critical challenges, including resource limitations, data security, energy efficiency, and interoperability across heterogeneous systems. In addition to exploring technical foundations, the article highlights real-world applications and use cases that demonstrate the practical value of AI at the edge. Finally, it discusses future directions, emphasizing the importance of adaptive systems, sustainable design, cross-layer optimization, and the development of standardized platforms to enable scalable, intelligent edge deployments. This work aims to serve as a foundational reference for researchers, engineers, and stakeholders seeking to understand the landscape, hurdles, and opportunities in the advancement of AI-driven edge computing systems.
| Origine | Publication financée par une institution |
|---|---|
| Licence |