An Architecture Integrity Simulation Evaluation Method for an Autonomous Transportation System Based on an Information-Triggered Collaboration Mechanism

Abstract

Autonomous transportation systems (ATSs) focus on traffic services and have become more capable of handling stochastic traffic scenarios. The preliminary work for designing and constructing ATS management and control systems is to develop an instructive and systematic architecture. However, the construction process of a conventional traffic system generally follows a function- and demand-oriented way. Thus, few studies have addressed architecture integrity evaluation and verification. In this work, we propose a collaboration mechanism-based approach to ATS architecture integrity evaluation. First, to provide a clear description of ATSs, ATS theory was systematically introduced through the generation, definition, and property of its basic elements, and then the relations between each element were analyzed based on the concept of microservices. Second, a collaboration mechanism on physical architecture was proposed to realize the adaptive construction of the architecture. Additionally, to fulfill the evaluation of architecture integrity, a contrast scheme combined with system function and information flow (IF) was determined, and accordingly, two independent ways were determined to obtain both the reference and calculated results. Finally, the scenario of automatic vehicles passing through an intersection was illustrated to demonstrate the feasibility of this static simulation evaluation method for the scenario architecture, and achieved a 100% function integrity rate and 93.3% IF integrity rate. A successful application in the case study shows that the proposed method could be extended to more general traffic scenarios.