Authors: Sukhdev Verma, Rakesh Tiwari, Sandeep Patil³, Dinesh Chauhan Abstract: Industrial machines are becoming more complex and distributed in modern smart manufacturing environments. Continuous monitoring of machine health is required to reduce unexpected downtime, improve productivity, and enhance operational safety. Traditional cloud-based monitoring approaches often suffer from latency, bandwidth limitations, and data privacy issues. Edge-AI embedded systems offer a promising solution by integrating artificial intelligence algorithms directly into embedded hardware located near machines. These systems enable real-time condition monitoring, anomaly detection, and predictive maintenance with minimal communication overhead. This paper reviews the architecture, components, algorithms, and applications of Edge-AI embedded systems for machine monitoring. The study also discusses recent advances in TinyML, edge sensors, and industrial IoT platforms that support machine health assessment. Challenges such as resource constraints, model optimization, and security concerns are highlighted along with future research directions. The results show that Edge-AI monitoring can significantly improve machine reliability and reduce maintenance cost in Industry 4.0 environments. Keywords: Edge AI, Embedded systems, Machine monitoring, Predictive maintenance, Industrial IoT, TinyML
Full Issue
| View or download the full issue | PDF 37-54 |