一句话：提出LeGEND，用两阶段LLM将自然语言功能场景转为形式化逻辑场景，实现多样且关键的ADS测试场景生成。
自动驾驶系统安全性至关重要，现有测试方法侧重场景关键性，却忽视场景多样性。为此，论文提出LeGEND，一种自顶向下的场景生成方法：从抽象的功能场景出发，逐步细化为逻辑场景和具体场景，从而在功能层面控制多样性。由于功能场景多以自然语言（如事故报告）记录，难以被计算机精确处理，LeGEND利用大语言模型（LLM）进行转换。为减少无关信息干扰，设计两阶段转换并引入中间语言：一个LLM从功能场景抽取信息，另一个将抽取信息转为形式化逻辑场景。在百度Apollo上的实验表明，LeGEND能有效识别关键场景，且在生成场景的多样性上显著优于基线方法，同时验证了两阶段框架和LLM的准确性。
Autonomous driving systems (ADS) are safety-critical and require comprehensive testing before their deployment on public roads. While existing testing approaches primarily aim at the criticality of scenarios, they often overlook the diversity of the generated scenarios that is also important to reflect system defects in different aspects. To bridge the gap, we propose LeGEND, that features a top-down fashion of scenario generation: it starts with abstract functional scenarios, and then steps downwards to logical and concrete scenarios, such that scenario diversity can be controlled at the functional level. However, unlike logical scenarios that can be formally described, functional scenarios are often documented in natural languages (e.g., accident reports) and thus cannot be precisely parsed and processed by computers. To tackle that issue, LeGEND leverages the recent advances of large language models (LLMs) to transform textual functional scenarios to formal logical scenarios. To mitigate the distraction of useless information in functional scenario description, we devise a two-phase transformation that features the use of an intermediate language; consequently, we adopt two LLMs in LeGEND, one for extracting information from functional scenarios, the other for converting the extracted information to formal logical scenarios. We experimentally evaluate LeGEND on Apollo, an industry-grade ADS from Baidu. Evaluation results show that LeGEND can effectively identify critical scenarios, and compared to baseline approaches, LeGEND exhibits evident superiority in diversity of generated scenarios. Moreover, we also demonstrate the advantages of our two-phase transformation framework, and the accuracy of the adopted LLMs.CCS CONCEPTS• Software and its engineering → Software testing and debugging; Search-based software engineering.
DOI 原文 · arXiv · PDF（开放获取） · DBLP
@inproceedings{abs-2409-10066,
title = {LeGEND: A Top-Down Approach to Scenario Generation of Autonomous Driving Systems Assisted by Large Language Models},
author = {Shuncheng Tang and Zhenya Zhang and Jixiang Zhou and Lei Lei and Yuan Zhou and Yinxing Xue},
booktitle = {ASE},
year = {2024},
doi = {10.1145/3691620.3695520}
}