Developing a Combined Surrogate Safety Measure for Rear-End Collisions by Fuzzy Inference System

This paper aims to develop a method to detect dangerous situations for each vehicle based on microscopic traffic data for intelligent vehicles. Here, surrogate safety measures (SSMs) would be applied. Each SSM has unique characteristics and if we could use the advantages of different SSM simultaneously, then the efficiency of intelligent vehicles might be increased. For this purpose, Fuzzy Inference System (FIS) is applied to develop a Combined Surrogate Safety Measure (CSSM). In order to avoid complicating the issue, only rear-end collisions are considered. Microscopic traffic data collected in Modares highway of Tehran is used to develop FIS. Finally, the CSSM results are compared by each SSM statistically. Based on the results, it can be declared that FIS can be helpful to calculate the rear-end collision probability by using different SSMs. This achievement can be useful in promoting the efficiency of autonomous vehicles.