AI-driven situational awareness for smart transportation systems

The research on Smart Transportation has been attracting much interest in recent years because of its potential to revolutionize the way transport occurs. Situational Awareness is one critical part of Smart Transportation that forms a perception of and comprehension of the current transportation environment. Applying Situational Awareness to Smart Transportation presents significant challenges, including heterogeneous data sources, real-time data processing demands, and the need for accurate and timely decision-making. This paper proposes an AI-based Smart Transportation Situational Awareness System (AI-STSAS) to address these challenges. The proposed approach utilizes image processing techniques to analyze transportation environments and continuously update situational information. The framework uses image processing techniques to extract visual features which include lane boundaries and object regions. AI-STSAS uses these visual features to make intelligent decisions and analyze traffic patterns and interpret real-time situations. A curve detection method and a situational warning mechanism are implemented to enhance driving safety. Simulation experiments conducted using the Waymo Open Dataset and the CARLA open-source simulator demonstrate that the proposed AI-STSAS achieves high accuracy in detection and classification and environmental conditions. The experimental results show that AI-STSAS achieves better performance than baseline models which include Random Forest (RF) because it provides a 10.9% boost in computational efficiency and a 12.9% increase in collision avoidance rates. The results demonstrate that the proposed system successfully improves both transportation safety and operational performance.