Semantic-Enhanced Digital Twin for Industrial Working Environments

Abstract

Real-time data from diverse Internet of Things (IoT) sensors (such as cameras, temperature, light, and air quality sensors) is essential for monitoring smart manufacturing environments. However, efficiently perceiving, integrating, and interpreting this data remains a challenge, as it involves dealing with heterogeneous data formats, ensuring data accuracy, and providing real-time analytics. This paper proposes a semantic-enhanced Digital Twin (DT) to address these complexities and aims to offer a comprehensive view of industrial working environments. The paper first presents a conceptual overview of the semantic-enhanced DT architecture, followed by a detailed description of the system architecture, encompassing edge, cloud, and interface modules. Additionally, the implementation of the entire system is presented. The results demonstrate the feasibility of the proposed DT, showing its potential for deployment in real-world scenarios.