arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2605.06264cs.LG
收藏

属性能否预测风险?从多视图属性到端到端自动驾驶中的规划风险信号

Can Attribution Predict Risk? From Multi-View Attribution to Planning Risk Signals in End-to-End Autonomous Driving

Sun Yat-sen University(中山大学);University of Chinese Academy of Sciences(中国科学院大学);Nanyang Technological University(南洋理工大学)

Le Yang, Haijun Liu, Jiawei Liang, ShangQuan Sun, Xiaochun Cao

AI总结:

本文提出了一种层次化属性框架,用于端到端规划中的风险预测,通过提取三种属性统计作为预测信号,提升了对轨迹误差和碰撞检测的预测能力。

摘要:

端到端自动驾驶模型通过多视图输入生成未来轨迹,提高了系统整合性,但引入了决策不透明和难以定位的风险。现有方法要么依赖辅助监控模型,要么生成文本解释,但与规划过程脱节,无法揭示轨迹生成的视觉证据。属性提供了一种直接替代方案,但规划不同于图像分类,因为它使用六视图相机图像作为输入,并预测连续多步轨迹,需要属性捕捉关键视图和区域及其对输出的影响。此外,属性映射是否能支持风险识别仍不明确。为此,我们提出了一种端到端规划的层次化属性框架。具体而言,使用L2一致性与原始轨迹作为目标,设计了从粗到细的区域属性策略,搜索全六视图输入中的候选区域并细化属性。我们进一步提取了三种属性统计作为规划风险的预测信号,包括属性熵以衡量规划器对联合视觉空间的依赖集中程度,单相机空间方差以表征属性在每个视图中的分布范围,以及跨相机基尼系数以量化属性在六个相机之间的分布不均程度。在BridgeAD、UniAD和GenAD上的实验表明,这些统计量与规划风险相关,实现了与轨迹误差的斯皮尔曼相关系数为0.30±0.07,碰撞检测的AUROC为0.77±0.04。该信号在未见场景中泛化良好,降解 negligible,且在替代属性基线下保持稳定。

英文摘要:

End-to-end autonomous driving models generate future trajectories from multi-view inputs, improving system integration but introducing opaque decisions and hard-to-localize risks. Existing methods either rely on auxiliary monitoring models or generate textual explanations, but are decoupled from the planning process and fail to reveal the visual evidence underlying trajectory generation. While attribution offers a direct alternative, planning differs from image classification by taking six-view camera images as input and predicting continuous multi-step trajectories, requiring attribution to capture both critical views and regions and their influence on outputs. Moreover, whether attribution maps can support risk identification remains underexplored. To address this, we propose a hierarchical attribution framework for end-to-end planning. Specifically, using L2 consistency with the original trajectory as the objective, we design a coarse-to-fine region attribution strategy that searches candidate regions across the full six-view input and refines attribution within them. We further extract three attribution statistics as predictive signals for planning risk, including attribution entropy to measure how concentrated the planner's reliance is over the joint visual space, within-camera spatial variance to characterize how spread out the attribution is within each view, and cross-camera Gini coefficient to quantify how unevenly attribution is distributed across the six cameras. Experiments on BridgeAD, UniAD, and GenAD show that these statistics correlate with planning risk, achieving Spearman correlations of $0.30 \pm 0.07$ with trajectory error and AUROC of $0.77 \pm 0.04$ for collision detection. The signal generalizes to held-out scenes with negligible degradation and remains stable under an alternative attribution baseline.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑