跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.20765 · v1

STA: A Split Temporal Adversary Method for Evaluating Robustness of Multimodal RL-based Driving Policies

首次提交 2026-11-16 17:16
版本更新 2026-11-16 17:16
雷达观测 2026-11-17 13:07
arXiv ↗PDF ↗

摘要

Multimodal reinforcement learning (MRL) offers a promising paradigm for autonomous driving by jointly leveraging complementary sensor observations to learn effective decision-making policies. However, its effectiveness critically depends on temporal alignment across modalities, making temporal inconsistency a fundamental vulnerability. Even subtle misalignment or latency can corrupt fused representations and ultimately lead to decision failures. Existing studies have explored adversarial attacks to induce cross-modal temporal inconsistencies, thereby enabling systematic evaluation of the temporal robustness of MRL policies. However, existing methods remain limited in generality, realism, and adaptivity. To address these limitations, we propose the Split Temporal Adversary (STA), which formulates multimodal temporal attacks as a unified decision process that jointly optimizes attack timing and modality selection under realistic policy-aware and budget-constrained settings. STA includes two mechanisms: a) a unified multimodal temporal perturbation mechanism that delays observations via historical states, removing the need for modality-specific noise; b) a split multimodal adversarial decision mechanism that jointly optimizes attack timing and modality selection through adaptive optimization while leveraging the victim policy's clean actions to identify sensitive decision patterns. We provide a theoretical analysis of the induced distribution shift and a modality selection principle. Experiments in the CARLA simulator show that STA more than doubles the collision rate of the strongest baseline against PPO while using 17.8\% fewer attack attempts, and that it generalizes to a VLM-based driving policy. These results reveal temporal consistency as a critical vulnerability, highlighting temporal robustness as essential for safe autonomous driving.

作者备注:10 pages, 6 figures, submitted for publication

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-16 17:16 · 最新观测版本

STA: A Split Temporal Adversary Method for Evaluating Robustness of Multimodal RL-based Driving Policies

Multimodal reinforcement learning (MRL) offers a promising paradigm for autonomous driving by jointly leveraging complementary sensor observations to learn effective decision-making policies. However, its effectiveness critically depends on temporal alignment across modalities, making temporal inconsistency a fundamental vulnerability. Even subtle misalignment or latency can corrupt fused representations and ultimately lead to decision failures. Existing studies have explored adversarial attacks to induce cross-modal temporal inconsistencies, thereby enabling systematic evaluation of the temporal robustness of MRL policies. However, existing methods remain limited in generality, realism, and adaptivity. To address these limitations, we propose the Split Temporal Adversary (STA), which formulates multimodal temporal attacks as a unified decision process that jointly optimizes attack timing and modality selection under realistic policy-aware and budget-constrained settings. STA includes two mechanisms: a) a unified multimodal temporal perturbation mechanism that delays observations via historical states, removing the need for modality-specific noise; b) a split multimodal adversarial decision mechanism that jointly optimizes attack timing and modality selection through adaptive optimization while leveraging the victim policy's clean actions to identify sensitive decision patterns. We provide a theoretical analysis of the induced distribution shift and a modality selection principle. Experiments in the CARLA simulator show that STA more than doubles the collision rate of the strongest baseline against PPO while using 17.8\% fewer attack attempts, and that it generalizes to a VLM-based driving policy. These results reveal temporal consistency as a critical vulnerability, highlighting temporal robustness as essential for safe autonomous driving.

查看此版本解读 →查看此版本原文 ↗