一个扩散模型，两种角色：闭环仿真中的引导式轨迹规划与安全关键场景生成
One Diffusion Model, Two Roles: Guided Trajectory Planning and Safety-Critical Scenario Generation in Closed-Loop Simulation
浏览论文内容
中文总结 AI 辅助
该研究提出单个扩散交通模型可同时承担自动驾驶开发中轨迹规划与安全关键场景生成两种角色，通过SSDS扩散-Transformer解码器和DAPSE方案提升规划性能，生成的场景可评估规划器鲁棒性并暴露其故障模式。
中文摘要 AI 辅助
扩散概率模型能够捕捉驾驶场景中联合未来轨迹的多模态、富含交互的分布。本文展示了单个预训练的扩散交通模型可在自动驾驶开发循环中承担两种互补角色：作为自车运动规划器，以及作为用于规划器压力测试的可控安全关键场景生成器。在规划方面，我们引入了单流双流（SSDS）扩散-Transformer解码器，其通过联合注意力而非后期交叉注意力融合场景上下文，提升了在nuPlan上的闭环性能。我们进一步提出了带能量的解耦退火后验采样（DAPSE），这是一种无需训练的引导方案，可在干净样本级别注入任意能量函数，避免了一阶近似误差且无需辅助网络。除规划外，我们将同一扩散模型用作可控场景生成器，以创建用于闭环评估的现实长尾驾驶交互。通过推理时引导，选定智能体被导向安全关键行为，包括激进加塞、前车制动及纵向-横向组合交互，同时保留现实交通行为。在使用独立黑盒规划器的闭环nuPlan仿真中评估，生成的场景暴露了在标准基准下仍隐藏的故障模式。尽管基于SSDS的规划器实现了更强的标称性能，但其在这些具有挑战性的场景下经历了更大的性能下降，表明基准优势不一定转化为鲁棒性。这些结果证明，单个学习到的交通先验可同时改进运动规划，并为系统的规划器鲁棒性评估提供现实框架。
英文摘要
Diffusion probabilistic models can capture the multi-modal, interaction-rich distribution of joint future trajectories in driving scenes. We show that a single pretrained diffusion traffic model can serve two complementary roles in the autonomous driving development loop: as an ego motion planner, and as a controllable generator of safety-critical scenarios for stress-testing the planners. On the planning side, we introduce a Single-Stream Dual-Stream (SSDS) diffusion-transformer decoder that fuses scene context via joint attention rather than late cross-attention, improving closed-loop performance on nuPlan. We further propose Decoupled Annealing Posterior Sampling with Energy (DAPSE), a training-free guidance scheme that injects arbitrary energy functions at the clean-sample level, avoiding the first-order approximation errors while requiring no auxiliary networks. Beyond planning, we leverage the same diffusion model as a controllable scenario generator to create realistic long-tail driving interactions for closed-loop evaluation. Through inference-time guidance, selected agents are steered toward safety-critical behaviors, including aggressive cut-ins, lead-vehicle braking, and combined longitudinal-lateral interactions, while preserving realistic traffic behaviors. Evaluated in closed-loop nuPlan simulations with independent black-box planners, the generated scenarios expose failure modes that remain hidden under standard benchmarks. Although the SSDS-based planner achieves stronger nominal performance, it experiences larger degradation under these challenging scenarios, demonstrating that benchmark superiority does not necessarily translate to robustness. These results demonstrate that a single learned traffic prior can simultaneously improve motion planning and provide a realistic framework for systematic planner robustness evaluation.
发表机构
- KTH Royal Institute of Technology（皇家理工学院）
- Zenseact AB（Zenseact公司）
- Chalmers University of Technology（查尔姆斯理工大学）
机构由 AI 辅助整理，请以论文原文为准。
