arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2401.07063cs.SE
收藏

ACAV:自动驾驶车辆事故记录自动因果分析框架

ACAV: A Framework for Automatic Causality Analysis in Autonomous Vehicle Accident Recordings

Huijia Sun, Christopher M. Poskitt, Yang Sun, Jun Sun, Yuqi Chen

AI总结:

针对现有自动驾驶测试方法仅能发现违规却无法定位致因的问题,提出两阶段自动化因果分析框架ACAV,经Apollo系统多组事故记录验证,可高效识别不同故障引发事故的因果事件。

摘要:

自动驾驶车辆(AVs)的快速发展让无人驾驶未来的前景比以往任何时候都更接近现实。然而,近期发生的致死事故凸显了通过大规模测试开展安全验证的重要性。已有多种方法借助高保真模拟器全自动实现这一目标,即生成多样化的驾驶场景,并依据不同的测试判定规则评估自动驾驶系统(ADSs)。这些方法虽能有效发现违规行为,却无法识别导致违规的决策与行动——而这类信息对提升ADS的安全性至关重要。为解决这一难题,我们提出ACAV,这是一个专为AV事故记录开展因果分析设计的自动化框架,分为两个阶段。第一阶段,我们采用基于ADS模块间交互消息的特征提取方案,并通过加权投票法剔除事故记录中与事故无关的帧。第二阶段,我们利用安全规范识别安全关键帧,并将我们的因果分析工具CAT应用于站点-时间图,推导因果事件。我们在Apollo ADS上对ACAV进行了评估,结果显示,在由AV测试引擎生成的110条事故记录中,ACAV能够识别出5种不同类型的因果事件,覆盖率达93.64%。我们进一步在1206条从注入特定故障的Apollo版本中收集的事故记录上评估了ACAV,结果表明,对于由预测错误引发的事故,ACAV能正确识别其中96.44%的因果事件;对于由规划错误引发的事故,这一比例为85.73%。

英文摘要:

The rapid progress of autonomous vehicles~(AVs) has brought the prospect of a driverless future closer than ever. Recent fatalities, however, have emphasized the importance of safety validation through large-scale testing. Multiple approaches achieve this fully automatically using high-fidelity simulators, i.e., by generating diverse driving scenarios and evaluating autonomous driving systems~(ADSs) against different test oracles. While effective at finding violations, these approaches do not identify the decisions and actions that \emph{caused} them -- information that is critical for improving the safety of ADSs. To address this challenge, we propose ACAV, an automated framework designed to conduct causality analysis for AV accident recordings in two stages. First, we apply feature extraction schemas based on the messages exchanged between ADS modules, and use a weighted voting method to discard frames of the recording unrelated to the accident. Second, we use safety specifications to identify safety-critical frames and deduce causal events by applying CAT -- our causal analysis tool -- to a station-time graph. We evaluate ACAV on the Apollo ADS, finding that it can identify five distinct types of causal events in 93.64% of 110 accident recordings generated by an AV testing engine. We further evaluated ACAV on 1206 accident recordings collected from versions of Apollo injected with specific faults, finding that it can correctly identify causal events in 96.44% of the accidents triggered by prediction errors, and 85.73% of the accidents triggered by planning errors.

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。

↑