Knowledge Graph-based Network Analysis on the Elements of Autonomous Transportation System

Abstract

Autonomous transportation system (ATS) is a further development on the basis of intelligent transportation system with strong ability of self-organizing operation and autonomous service. In order to intuitively display the composition and correlation of ATS, the characteristics of elements and their potential laws are deeply explored. In this paper, the entity categories are defined based on five traffic elements: technology, demand, service, function and component, and the correlations are established through their attributes. With the help of Neo4j graph database, the structured data information is stored to build a network knowledge graph of transportation system. Accordingly, the co-occurrence analysis was used to identify the potential correlation of similar elements and calculate their correlation strength. Meanwhile, the key degree of each element in the transportation system is studied by combining the correlation indicators and statistics of the correlation relationship of different elements. Finally, the “V2X based Vehicle Collision Warning for Autonomous Vehicles” scenario is taken as an example for empirical analysis. Through the knowledge graph analysis, the correlations between various elements of the autonomous transportation system are visualized, and the linkage mechanism is deeply identified. Also, this work can provide theoretical support for the construction and improvement of the system architecture in the future.