A Survey on Collaborative Collision Warning of Connected Vehicles at Signal-Free Intersections^*

To clarify the limitation and development direction of Collaborative Collision Warning (CCW) research at signal-free intersections, the survey is systematically combed through the investigation and summary of this research field. It compares and contrasts the existing intersection traffic conflict detection methods from five aspects, summarizes the applicability of different warning levels, mechanisms, and modes, summarizes the evaluation indexes related to the effectiveness and safety benefit of CCW, analyzes the impact of driver uncertainty on CCW and its optimization ideas, and envisions the future development direction of CCW. The research results show that the driving safety field model is a feasible solution to the problems of characterizing the comprehensive traffic risk and measuring the conflict severity in the existing CCW conflict detection methods, but the parameters, indicators, and thresholds related to traffic environment and conflict risk still need further study. The driver’s personality and psychological characteristics are one of the key challenges in the effectiveness and reliability design of CCW applications. Accurate modeling and prediction of driver’s driving characteristics in CCW application scenarios, online identification of driver’s driving characteristics, and adaptive construction of CCW warning mode are possible solutions. A comprehensive evaluation mechanism of driving safety, traffic efficiency, and driving comfort is needed to evaluate CCW’s function and performance objectively.