Investigating the synergistic effects of temporal knowledge graphs and digital twin technologies to enhance situational awareness decision-making in smart cities

The growing rate of urbanization accompanied by population growth has convoluted the problems identical with urban road traffic, including frequent accidents, traffic jams, and infractions, rendering conventional traffic control strategies insufficient. To address these issues, this study explores the integration of advanced technologies, such as temporal knowledge graphs (TKG) and digital twin (DT) technologies, which offer a promising solution for enhancing situational awareness and decision-making in urban traffic management. In this study, a novel Efficient Tunicate Swarm Optimized Region-Based Convolutional Neural Network (ETSO-RCNN) is proposed for object detection and tracking of vehicles. Traffic data is gathered from various urban traffic cameras, primarily through video feeds, which document traffic flow patterns, congestion levels, and vehicle movements. The data was preprocessed using data cleaning, and noise was removed from the data using a Kalman filter (KF). Scale-Invariant Feature Transform (SIFT) was employed for feature extraction. The information is utilized in a DT model to visualize urban traffic flow, utilizing TKG for predictive insights and proactive decision-making. The results indicate the R-CNN is outperformed in accurate vehicle detection and recognition compared to other traditional algorithms. By anticipating possible traffic problems, allocating resources more effectively, and enabling real-time modifications to traffic management plans, the combined TKG-DT strategy improves situational awareness. The study highlights how integrating TKGs and DTs into smart city infrastructures can have a revolutionary effect by providing a scalable and flexible way to manage intricate urban road networks.