Xie, Yizhou (Shanghai Jiao Tong University), Zhang, Yong (Shanghai Jiao Tong University), Yin, Chengliang (Shanghai Jiao Tong University)

Generating Long-Term Vehicle-Vehicle Scenarios with High-Density Threatening Maneuvers in the Given Test Site

Scheduled for presentation during the Regular Session "S27a-Safety and Risk Assessment for Autonomous Driving Systems" (TH-LM-T27), Thursday, November 20, 2025, 12:10−12:30, Broadbeach 3

2025 IEEE 28th International Conference on Intelligent Transportation Systems (ITSC), November 18-21, 2025, Gold Coast, Australia

This information is tentative and subject to change. Compiled on October 18, 2025

Abstract

Testing an autonomous vehicle (AV) in the actual test field or test road is the most reliable verification approach, and in recent years, sensor simulation techniques that can apply virtual scenarios to test real AV have been developed, which significantly reduce the cost of scenario-based tests in actual test sites. However, existing researches still lack the studies in continuous scenario generation in test sites, especially in the scenario arrangement considering the efficiency and physical constraint in the real world. In this paper, we propose a long-term scenario generation framework, which is able to arrange high-density and diversified threatening sub-scenarios within the given road network. We carry out a real-time test to verify the proposed framework. Test results indicate that the generation through our framework involves threatening maneuvers of participant vehicles in the scene continuously, and efficiently covers different types of logical scenarios.