Infrastructure-Scheduled In-Vehicle Auditory Signals for Manual Driving Coordination in V2I Mixed Traffic

Abstract

Traffic systems are undergoing a major transformation with the advent of autonomous driving, but the transition to mixed traffic where automatically driven (AD) and human-driven (HD) vehicles coexist remains unavoidable. Vehicle-to-infrastructure (V2I) technologies have emerged as a promising coordinator between autonomous and human-driven vehicles, but existing works face limitations in influencing HD drivers. Taking a cue from in-vehicle auditory signals in advanced driver-assistance systems (ADAS), we propose a framework in which infrastructure communicates its trajectory planning to HD vehicles via designed in-vehicle auditory signals. We validate this framework through two case studies: right-turn bicycle collision avoidance and merging speed adjustment, where scenario-specific auditory signals were designed and compared with conventional signals. Simulation experiments with twelve participants showed that our signals enabled drivers to brake 1.1 seconds earlier than with conventional beeps in right-turn scenarios without reducing traffic efficiency, and improved accuracy and robustness of speed adjustment across diverse merging conditions. These results demonstrate the feasibility of the proposed platform and its potential as a quantitative environment for infrastructure-based interventions in mixed traffic.