一句话：提出 ScenarioFuzz：利用地图路网与历史测试数据、结合 GNN 筛选高风险场景种子的 ADS 场景模糊测试方法。
自动驾驶系统安全验证中，场景模糊测试常依赖预定义场景，难以高效发现失效。本文提出 ScenarioFuzz，从 OPENDRIVE 等地图路网提取信息构建基础场景种子语料，并在无初始场景时提供模糊测试边界；结合专用变异器、变异技术与图神经网络模型，基于历史测试数据预测并过滤高风险场景种子，以优化模糊测试过程。作者还提出自监督碰撞轨迹聚类方法，归纳 54 类易诱发 ADS 故障的高风险场景。实验表明，相较其他方法，时间开销平均降低 60.3%，单位时间发现的错误场景数增加 103%，并在六个被测系统中发现 58 个 bug。
As autonomous driving systems (ADS) advance towards higher levels of autonomy, orchestrating their safety verification becomes increasingly intricate. This paper unveils ScenarioFuzz, a pioneering scenario-based fuzz testing methodology. Designed like a choreographer who understands the past performances, it uncovers vulnerabilities in ADS without the crutch of predefined scenarios. Leveraging map road networks, such as OPENDRIVE, we extract essential data to form a foundational scenario seed corpus. This corpus, enriched with pertinent information, provides the necessary boundaries for fuzz testing in the absence of starting scenarios. Our approach integrates specialized mutators and mutation techniques, combined with a graph neural network model, to predict and filter out high-risk scenario seeds, optimizing the fuzzing process using historical test data. Compared to other methods, our approach reduces the time cost by an average of 60.3%, while the number of error scenarios discovered per unit of time increases by 103%. Furthermore, we propose a self-supervised collision trajectory clustering method, which aids in identifying and summarizing 54 high-risk scenario categories prone to inducing ADS faults. Our experiments have successfully uncovered 58 bugs across six tested systems, emphasizing the critical safety concerns of ADS.
DOI 原文 ·
@inproceedings{paperbot8596,
title = {Dance of the ADS: Orchestrating Failures through Historically-Informed Scenario Fuzzing},
author = {Tong Wang and Taotao Gu and Huan Deng and Hu Li and Xiaohui Kuang and Gang Zhao},
booktitle = {ISSTA},
year = {2024},
doi = {10.1145/3650212.3680344}
}