Construction of a Joint Simulation Platform Using SUMO, CARLA, and Autoware and Its Application in Multi-Vehicle Interactions at Unsignalized Intersections

This study aims to construct a joint simulation platform based on SUMO, CARLA, and Autoware, and apply the platform to the investigation of multi-vehicle game scenarios at unsignalized intersections. By leveraging the strengths of these three open-source tools, we created an efficient environment for modeling complex traffic scenarios and evaluating autonomous driving algorithms with a low-code approach. The research primarily focuses on reproducing deadlock situations encountered by Robotaxis during public road tests and assessing the impact of vehicle-to-infrastructure (V2X) cooperation on traffic efficiency through comparative experiments. The results demonstrate that the proposed platform effectively simulates multi-vehicle interactions, serving as a valuable tool for optimizing autonomous driving systems.