Current Machine Learning Applications in Road Safety Monitoring: A Comprehensive Review

As societies around the world face increasing traffic challenges and encounter various road conditions, we explore in this work the potential of Machine Learning and Deep Learning methodologies in assisting with road safety decision making across its five fundamental pillars. Road safety management, safer roads, safer vehicles, safer road users, and post-accident response. These pillars form a comprehensive framework for improving road safety with the objective of reducing trafficrelated fatalities and injuries by addressing five different aspects of road transport. We explore the literature to examine the diverse applications of Machine Learning across these five critical dimensions. The review illuminates the versatility of Machine Learning techniques, ranging from traditional signal processing to advanced deep learning approaches. The findings underscore the profound impact of Machine Learning on road safety, offering actionable insights for accident prediction, infrastructure optimization, and a nuanced understanding of user behaviors. This study serves as an instructive roadmap, emphasizing the indispensable role of Machine Learning in shaping safer, adaptive, and efficient road environments.