动态生成非玩家车辆行为，更高效发现自动驾驶系统真实缺陷
DynNPC: Finding More Violations Induced by ADS in Simulation Testing through Dynamic NPC Behavior Generation
- NPC车辆在仿真中实时根据交通信号和自车行为动态调整驾驶策略
- 相比现有方法，发现的违规场景中由自车系统引发的比例提升42%
- 适合自动驾驶测试团队快速定位真实系统漏洞
近期多项仿真测试方法被提出，用于为自动驾驶系统（ADS）生成多样化的驾驶场景。然而，以往方法中非玩家车辆（NPC）的行为在仿真前预先设定并迭代变异，忽略了交通信号及自车实时行为，导致大量违规由不现实的NPC行为引发，无法揭示ADS的真实缺陷。此外，NPC行为在迭代变异中的庞大搜索空间也限制了效率。为此，本文提出新型基于场景的测试框架DynNPC，使NPC车辆在仿真执行过程中依据交通信号与自车实时行为，动态生成不同驾驶策略。评估表明，与现有最先进方法相比，DynNPC在发现由自车系统引发的违规场景方面更具有效性和效率。
原文摘要 · Abstract (English)
Recently, a number of simulation testing approaches have been proposed to generate diverse driving scenarios for autonomous driving systems (ADSs) testing. However, the behaviors of NPC vehicles in these scenarios generated by previous approaches are predefined and mutated before simulation execution, ignoring traffic signals and the behaviors of the Ego vehicle. Thus, a large number of the violations they found are induced by unrealistic behaviors of NPC vehicles, revealing no bugs of ADSs. Besides, the vast scenario search space of NPC behaviors during the iterative mutations limits the efficiency of previous approaches. To address these limitations, we propose a novel scenario-based testing framework, DynNPC, to generate more violation scenarios induced by the ADS. Specifically, DynNPC allows NPC vehicles to dynamically generate behaviors using different driving strategies during simulation execution based on traffic signals and the real-time behavior of the Ego vehicle. We compare DynNPC with state-of-the-art scenario-based testing approaches. Our evaluation has demonstrated the effectiveness and efficiency of DynNPC in finding more violation scenarios induced by the ADS.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。