发表机构
Tsinghua University; The Hong Kong University of Science and Technology (Guangzhou); Neolix（清华大学; 香港科技大学（广州）; 新石器无人车）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对端到端自动驾驶中稀疏场景表示导致候选生成质量低和排序不可靠的问题，提出RRDrive，引入风险感知占用表示，提升候选质量与排序，实验显示显著性能提升。
AI 中文摘要
多模态轨迹预测提高了端到端自动驾驶中的行为覆盖率，但现有方法仍受限于稀疏的场景表示。不完整的证据导致候选生成质量低下，以及在几何相似的轨迹之间排序不可靠。在基于寄存器的基线上，差和较差的候选占候选集的19.74%，而oracle最佳候选平均仅排在第33.9位。我们提出RRDrive，它引入风险感知占用作为密集、时间对齐且可轨迹查询的表示。其全局结构指导高质量的多模态生成，而候选条件风险查询支持细粒度选择。我们进一步构建了带有自动风险标注的RiskOcc4D-NAVSIM。RRDrive实现了所选轨迹的PDMS为0.951，相对于基线（0.937）相对提升1.5%，并将平均候选PDMS提高了7.7%。在具有挑战性的场景中，它将候选PDMS提高了30.2%，并将优秀候选之间的Spearman相关性从0.26提高到0.67，增加了0.41。为了超越这种oracle设置，我们进一步开发了一个外部RiskOcc预测器，这是一个直接从传感器输入估计风险感知占用的感知模块。竞争性的性能验证了该表示的可行性。
英文摘要
Multimodal trajectory prediction improves behavioral coverage in end-to-end autonomous driving, but existing methods remain limited by sparse scene representations. Incomplete evidence leads to low-quality candidate generation and unreliable ranking among geometrically similar trajectories. On a register-based baseline, bad and poor candidates constitute 19.74% of the candidate set, while the oracle-best candidate ranks only 33.9th on average. We propose RRDrive, which introduces risk-aware occupancy as a dense, temporally aligned, and trajectory-queryable representation. Its global structure guides high-quality multimodal generation, while candidate-conditioned risk queries support fine-grained selection. We further construct RiskOcc4D-NAVSIM with automatic risk annotations. RRDrive achieves a selected-trajectory PDMS of 0.951, representing a 1.5% relative improvement over the baseline (0.937), and improves the average candidate PDMS by 7.7%. In challenging scenes, it improves candidate PDMS by 30.2% and increases the Spearman correlation among good candidates by 0.41, from 0.26 to 0.67. To move beyond this oracle setting, we further develop an external RiskOcc predictor, a perception module that estimates risk-aware occupancy directly from sensor inputs. The competitive performance validates the representation's feasibility.
CommentsThis version of this research was completed in early 2026
