一句话：提出 LiRTest，通过增强 LiDAR 点云来自动化测试自动驾驶系统，缓解人工采集标注成本高的问题。
随着深度神经网络的发展，自动驾驶系统已应用于多种安全关键任务，但现实中仍发生事故甚至致命案例。其感知能力依赖复杂且难以解释的 DNN 模型，使传统软件测试技术难以直接适用于 ADS 感知；而现有测试方法多依赖人工采集与标注数据，过程耗时且成本极高。标题与摘要表明，LiRTest 旨在通过增强 LiDAR 点云来支持自动驾驶系统的自动化测试，降低对人工数据采集和标注的依赖。不过，摘要只阐述了研究动机与问题背景，未展开具体点云增强策略、测试流程或实验评估，因此无法从中报告关键结果。
With the tremendous advancement of Deep Neural Networks (DNNs), autonomous driving systems (ADS) have achieved significant development and been applied to assist in many safety-critical tasks. However, despite their spectacular progress, several real-world accidents involving autonomous cars even resulted in a fatality. While the high complexity and low interpretability of DNN models, which empowers the perception capability of ADS, make conventional testing techniques inapplicable for the perception of ADS, the existing testing techniques depending on manual data collection and labeling become time-consuming and prohibitively expensive.
DOI 原文 ·
@inproceedings{paperbot7438,
title = {LiRTest: augmenting LiDAR point clouds for automated testing of autonomous driving systems},
author = {An Guo and Yang Feng and Zhenyu Chen},
booktitle = {ISSTA},
year = {2022},
doi = {10.1145/3533767.3534397}
}