Development and safety evaluation of an adaptive personalized speed guidance system for on-ramp merging in highway service areas

Abstract

As autonomous driving and Vehicle-to-Everything (V2X) technologies evolve, the efficiency and safety of ramp merging in the highway service areas have become increasing critical. This study introduces a closed-loop feedback speed guidance system that accommodates individual driving styles, aiming to optimize merging behaviour, reduce traffic accidents, and enhance total traffic efficiency. The system dynamically adjusts the merging vehicle speeds by continuously monitoring their speed and location with variable steps to promote smoother merging. Moreover, this research also involves collecting naturalistic driving data from ramp merging scenarios, using the K-means clustering and point estimation method to recognize and analyse driving style characteristics, and integrating these styles into the developed closed-loop feedback speed guidance system. This approach results in personalized speed guidance curves tailored to different driving styles, facilitating more efficient mering. Additionally, the study conducts a Safety of the Intended Functionality (SOTIF) evaluation of this system using the System-Theoretic Process Analysis (STPA) method, which helps identify potential security risks and develop appropriate mitigation strategies to ensure the system’s safe and stable operation. The simulation results confirm that this innovative dynamic speed guidance system substantially improves traffic safety and efficiency in ramp merging areas.