BridgeGuard：基于扩散模型的自动驾驶的显式安全漂移
BridgeGuard: Explicit Safety Drift for Diffusion-based Autonomous Driving
浏览论文内容
中文总结 AI 辅助
针对扩散驾驶规划器在分布偏移下的不安全轨迹问题，提出BridgeGuard方法，通过安全约束的扩散规划提升了BridgeDrive和DiffusionDrive^geo在Bench2Drive上的驾驶性能，具备跨模型泛化能力。
中文摘要 AI 辅助
基于扩散模型的驾驶规划器可捕捉多样化行为，但在分布偏移下可能生成不安全轨迹。我们提出BridgeGuard，一种安全约束的扩散规划方法，在去噪过程中逐步强化约束项，引导中间轨迹向场景依赖的安全域靠近。修正操作在低维曲线空间中进行，以提升几何一致性。学习模块DistanceFieldNet从鸟瞰图特征预测随时间变化的距离场，对在专家轨迹外采样的查询点施加值和空间梯度监督，使该场学习到安全与不安全区域的信息。该学习场通过安全注入提供约束项，同时预训练的感知骨干网络和规划器保持冻结。我们还在理想化的连续时间桥梁场景中建立了终端安全的充分条件。在Bench2Drive上，BridgeGuard使BridgeDrive的驾驶得分/成功率从87.99%/74.99%提升至90.88%/76.36%，使DiffusionDrive^geo的驾驶得分/成功率从80.79%/58.18%提升至90.46%/74.09%，展现了跨模型泛化能力。
英文摘要
Diffusion-based driving planners capture diverse behaviors but can generate unsafe trajectories under distribution shift. We propose BridgeGuard, a safety-constrained diffusion planning method that progressively strengthens a constraint term during denoising to drive intermediate trajectories toward a scene-dependent safety domain. Corrections operate in a low-dimensional curve space, promoting geometric coherence. A learned module, DistanceFieldNet, predicts a time-dependent distance field from bird's-eye-view features. Value and spatial-gradient supervision at queries sampled beyond expert trajectories teaches this field about both safe and unsafe regions. The learned field supplies the constraint term through safety injection while the pretrained perception backbone and planner remain frozen. We further establish sufficient conditions for terminal safety in an idealized continuous-time bridge. On Bench2Drive, BridgeGuard improves driving score/success rate from 87.99/74.99% to 90.88/76.36% for BridgeDrive and from 80.79/58.18% to 90.46/74.09% for $\text{DiffusionDrive}^{\text{geo}}$, demonstrating cross-model generalization.
发表机构
- Bosch Research（博世研究院）
- Shanghai Jiao Tong University（上海交通大学）
机构由 AI 辅助整理，请以论文原文为准。
