Artificial intelligence techniques in road safety for sustainable mobility: an investigation into applications, challenges, and opportunities

Road safety is a paramount concern in transportation engineering, and leveraging artificial intelligence (AI) techniques has emerged as a promising approach to address this critical issue. This study investigates the applications, challenges, and opportunities associated with the utilization of AI in the context of road safety for sustainable mobility, with a particular focus on uncertainty and decision-making processes. This chapter explores a wide range of applications where AI techniques can significantly impact road safety for sustainable mobility, including machine learning (ML) and deep learning (DL) techniques. These applications encompass driver behavior analysis, accident prediction and prevention, traffic flow optimization, and intelligent transportation systems. By analyzing vast amounts of data, ML and DL algorithms can uncover hidden patterns and correlations, enabling proactive measures to mitigate road accidents. However, the presence of uncertainty within the data and decision-making processes poses significant challenges that need to be addressed. Uncertainty arises due to factors such as incomplete or noisy data, inherent variability in road conditions, and unpredictability of human behavior. Effectively managing uncertainty is crucial for developing robust and reliable ML and DL models in road safety applications. In addition to uncertainty, decision-making in road safety poses unique challenges. AI techniques must not only provide accurate predictions but also support transparent and interpretable decision-making processes. The implications of decision-making in road safety further emphasize the need for understanding the reasoning behind algorithmic outputs. The primary objective of this chapter is to examine how AI techniques can be leveraged to enhance the current state of data-driven analysis for road safety in transportation engineering. The study involves exploring conceptual frameworks, benefits, challenges, and existing approaches in this domain. This chapter provides an overview of the role of AI techniques in future data-driven road safety analysis for sustainable mobility. Additionally, it emphasizes and provides detailed insights into potential research opportunities in AI-assisted road safety.