Traffic State Information Extraction Methods Based on Granular Computing

Abstract

In order to extract traffic state information and provide decision support for traffic management, Granular computing theory was applied in traffic information processing. Traffic information granule and its granularity were defined, and then a methodology that provides a framework of traffic management and decision-making was presented based on GrC. A method was proposed for traffic state information granule construction based on vague sets, and then travel state identification model was proposed based on traffic state information granule similarity. The methods of traffic state information granule construction and their granularity were discussed based on a demonstration network in detail. The results show that the existing traffic information processing methods could be integrated based on GrC, and the proposed methodology can satisfy the demand of traffic management decision-making.