一句话：提出多目标搜索测试框架 MOSAT，用原子机动和 motif 模式生成多样对抗场景，发现 ADS 安全违规。
自动驾驶系统是安全攸关系统，部署前必须通过仿真测试充分暴露其安全违规。本文提出 MOSAT，一个基于多目标搜索的测试框架，用于构造多样且对抗的驾驶环境。MOSAT 基于原子驾驶机动引入 motif 模式，用其描述能有效挑战 ADS 的机动序列，并利用多目标遗传算法搜索对抗且多样的测试场景。为在长里程驾驶中全面测试 ADS，作者还设计了一种连续仿真测试技术，交替运行多个并行搜索过程生成的场景，持续对 ADS 施加不同扰动。在工业级平台百度 Apollo 上的实验表明，MOSAT 能有效生成安全攸关场景使 ADS 发生碰撞，短时间内暴露 11 类不同的安全违规，并在同一道路上比现有最先进技术多发现 6 类安全违规。
Autonomous Driving Systems (ADSs) are safety-critical systems, and safety violations of Autonomous Vehicles (AVs) in real traffic will cause huge losses. Therefore, ADSs must be fully tested before deployed on real world roads. Simulation testing is essential to find safety violations of ADS. This paper proposes MOSAT, a multi-objective search-based testing framework, which constructs diverse and adversarial driving environment to expose safety violations of ADSs. Specifically, based on atomic driving maneuvers, MOSAT introduces motif pattern, which describes a sequence of maneuvers that can challenge ADS effectively. MOSAT constructs test scenarios by atomic maneuvers and motif patterns, and uses multi-objective genetic algorithm to search for adversarial and diverse test scenarios. Moreover, in order to test the performance of ADS comprehensively during long-mile driving, we design a novel continuous simulation testing technique, which runs the scenarios generated by multiple parallel search processes alternately in the simulator and can continuously create different perturbations to ADS. We demonstrate MOSAT on an industrial-grade platform, Baidu Apollo, and the experimental results show that MOSAT can effectively generate safety-critical scenarios to crash ADSs and it exposes 11 distinct types of safety violations in a short period of time. It also outperforms state-of-the-art techniques by finding more 6 distinct safety violations on the same road.
DOI 原文 ·
@inproceedings{paperbot7068,
title = {MOSAT: finding safety violations of autonomous driving systems using multi-objective genetic algorithm},
author = {Haoxiang Tian and Yan Jiang and Guoquan Wu and Jiren Yan and Jun Wei and Wei Chen and Shuo Li and Dan Ye},
booktitle = {FSE},
year = {2022},
doi = {10.1145/3540250.3549100}
}