Surrogate Safety Measure

Surrogate safety measures (SSMs) are quantitative indicators designed to predict the likelihood of traffic accidents without relying on historical crash data, enabling proactive safety analysis. Current research focuses on developing more accurate and robust SSMs using machine learning techniques, such as deep learning models and Gaussian Process Regression, often incorporating probabilistic frameworks and considering diverse factors like driver behavior and pedestrian classifications. This work is crucial for improving real-time safety systems in various applications, from autonomous vehicle development to enhancing pedestrian safety at intersections, by providing more reliable and nuanced risk assessments.