当规划器遇上现实：学习型反应式交通智能体如何改变 nuPlan 基准测试
When Planners Meet Reality: How Learned, Reactive Traffic Agents Shift nuPlan Benchmarks
- Robert Bosch GmbH（罗伯特·博世有限公司）
- Institute for Neuro- and Bioinformatics, University of Lübeck（神经与生物医学研究所，吕贝克大学）
- Engineering Faculty, DHBW Stuttgart（工程学院，斯图加特DHBW）
- Department of Computer Science, University of Freiburg（计算机科学系，弗赖堡大学）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本研究将学习型反应式交通智能体SMART集成到nuPlan，发现基于IDM的仿真高估规划性能，而学习型规划器在交互场景表现更优，建议采用SMART反应式仿真作为新基准。
AI中文摘要：
闭环仿真中的规划器评估通常使用基于规则的交通智能体，其简单且被动的行为可能掩盖规划器的缺陷并导致排名偏差。广泛使用的IDM智能体仅跟随前车，无法对相邻车道上的车辆做出反应，从而阻碍了对复杂交互能力的测试。我们通过将最先进的学习型交通智能体模型SMART集成到nuPlan中来解决这一问题。因此，我们首次在更现实的条件下评估规划器，并量化当缩小仿真与现实的差距时结论如何变化。我们的分析涵盖了14个近期规划器和已建立的基线，结果表明基于IDM的仿真高估了规划性能：几乎所有分数都下降。相比之下，许多规划器的交互能力比之前假设的更好，甚至在多车道、交互密集的场景（如变道或转弯）中表现有所提升。在闭环中训练的方法展现出最佳且最稳定的驾驶性能。然而，在增强的边缘场景中达到极限时，所有学习型规划器都会突然退化，而基于规则的规划器则保持合理的基本行为。基于我们的结果，我们建议将SMART反应式仿真作为nuPlan中新的标准闭环基准，并在https://github.com/shgd95/InteractiveClosedLoop发布SMART智能体作为IDM的即插即用替代方案。
英文摘要：
Planner evaluation in closed-loop simulation often uses rule-based traffic agents, whose simplistic and passive behavior can hide planner deficiencies and bias rankings. Widely used IDM agents simply follow a lead vehicle and cannot react to vehicles in adjacent lanes, hindering tests of complex interaction capabilities. We address this issue by integrating the state-of-the-art learned traffic agent model SMART into nuPlan. Thus, we are the first to evaluate planners under more realistic conditions and quantify how conclusions shift when narrowing the sim-to-real gap. Our analysis covers 14 recent planners and established baselines and shows that IDM-based simulation overestimates planning performance: nearly all scores deteriorate. In contrast, many planners interact better than previously assumed and even improve in multi-lane, interaction-heavy scenarios like lane changes or turns. Methods trained in closed-loop demonstrate the best and most stable driving performance. However, when reaching their limits in augmented edge-case scenarios, all learned planners degrade abruptly, whereas rule-based planners maintain reasonable basic behavior. Based on our results, we suggest SMART-reactive simulation as a new standard closed-loop benchmark in nuPlan and release the SMART agents as a drop-in alternative to IDM at https://github.com/shgd95/InteractiveClosedLoop.
