Real-Time Infrastructure Monitoring System Using Digital Twin and AI-Based Computer Vision Technique

The research project is non-economic in nature, as the scientific and technological knowledge generated will be disseminated in line with public information objectives. The research activities carried out within the project fall under the category of industrial research. The evolution of digital twin technology, combined with IoT-based real-time monitoring, states a transformative approach to infrastructure management, especially in steel bridge. This study focuses on developing a real-time infrastructure monitoring system for existing steel bridges in Latvia. The aim is to detect structural changes in steel bridge, predict failures, and enable proactive maintenance through a continuous real-time AI-driven monitoring system. Sensors such as accelerometers, strain gauges, temperature probes and high-resolution cameras will be strategically installed to capture real-time data which will communicate through LoRaWAN to ensure minimal latency in data processing. A critical aspect of this study is the development of a digital twin, which serves as a dynamic, real-time virtual replica of the physical steel bridge in Latvia. This digital representation will be created using Building Information Modelling, integrating real-time sensor feeds for continuous monitoring. A dashboard will be developed to visualize the live status of the steel bridge behaviour, enabling engineers and decision-makers to monitor key parameters such as structural stress, environmental conditions, and potential anomalies etc. The system will be designed to provide automated alerts when abnormal conditions are detected, not only in real-time but foresee the future changes, allowing for immediate interference.