用扩散模型生成多条轨迹，让自动驾驶系统更懂不确定性。
Using Ensemble Diffusion to Estimate Uncertainty for End-to-End Autonomous Driving
- 将扩散模型与注意力池化结合，统一处理感知与轨迹规划。
- 单帧输入可生成128条候选轨迹，显式建模未来路径的不确定性。
- 无需额外模块，适配现有端到端系统，提升安全性与可解释性。
端到端自动驾驶规划系统在封闭仿真环境（如CARLA）中快速进步，但多数系统未将不确定性纳入规划，或依赖难以泛化的专用表示。本文提出EnDfuser，一种基于扩散模型的端到端驾驶系统。它通过单一扩散变压器模块，融合摄像头与激光雷达特征，并利用集成扩散生成128条候选轨迹，从单帧感知中捕捉复杂未来的不确定性分布。该方法使系统能以可视化方式理解多模态轨迹空间，进而设计简单安全规则，在LAV基准上提升驾驶得分1.7%。结果表明，将集成扩散作为传统点估计规划模块的即插即用替代方案，可有效支持不确定性感知的决策过程，对后验轨迹分布进行建模。
原文摘要 · Abstract (English)
End-to-end planning systems for autonomous driving are rapidly improving, especially in closed-loop simulation environments like CARLA. Many such driving systems either do not consider uncertainty as part of the plan itself or obtain it by using specialized representations that do not generalize. In this paper, we propose EnDfuser, an end-to-end driving system that uses a diffusion model as the trajectory planner. EnDfuser effectively leverages complex perception information like fused camera and LiDAR features, through combining attention pooling and trajectory planning into a single diffusion transformer module. Instead of committing to a single plan, EnDfuser produces a distribution of candidate trajectories (128 for our case) from a single perception frame through ensemble diffusion. By observing the full set of candidate trajectories, EnDfuser provides interpretability for uncertain, multimodal future trajectory spaces. Using this information we design a simplistic safety-rule that improves the system's driving score by 1.7% on the LAV benchmark. Our findings suggest that ensemble diffusion, used as a drop-in replacement for traditional point-estimate trajectory planning modules, can contribute to an uncertainty-aware decision making process in End-to-End driving policies by modeling the uncertainty of the posterior trajectory distribution.
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。