通过模块噪声攻克端到端自动驾驶技术

Lu Wang, Tianyuan Zhang, Yikai Han, Muyang Fang, Ting Jin, Jiaqi Kang
{"title":"通过模块噪声攻克端到端自动驾驶技术","authors":"Lu Wang, Tianyuan Zhang, Yikai Han, Muyang Fang, Ting Jin, Jiaqi Kang","doi":"arxiv-2409.07706","DOIUrl":null,"url":null,"abstract":"With recent breakthroughs in deep neural networks, numerous tasks within\nautonomous driving have exhibited remarkable performance. However, deep\nlearning models are susceptible to adversarial attacks, presenting significant\nsecurity risks to autonomous driving systems. Presently, end-to-end\narchitectures have emerged as the predominant solution for autonomous driving,\nowing to their collaborative nature across different tasks. Yet, the\nimplications of adversarial attacks on such models remain relatively\nunexplored. In this paper, we conduct comprehensive adversarial security\nresearch on the modular end-to-end autonomous driving model for the first time.\nWe thoroughly consider the potential vulnerabilities in the model inference\nprocess and design a universal attack scheme through module-wise noise\ninjection. We conduct large-scale experiments on the full-stack autonomous\ndriving model and demonstrate that our attack method outperforms previous\nattack methods. We trust that our research will offer fresh insights into\nensuring the safety and reliability of autonomous driving systems.","PeriodicalId":501301,"journal":{"name":"arXiv - CS - Machine Learning","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Attack End-to-End Autonomous Driving through Module-Wise Noise\",\"authors\":\"Lu Wang, Tianyuan Zhang, Yikai Han, Muyang Fang, Ting Jin, Jiaqi Kang\",\"doi\":\"arxiv-2409.07706\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"With recent breakthroughs in deep neural networks, numerous tasks within\\nautonomous driving have exhibited remarkable performance. However, deep\\nlearning models are susceptible to adversarial attacks, presenting significant\\nsecurity risks to autonomous driving systems. Presently, end-to-end\\narchitectures have emerged as the predominant solution for autonomous driving,\\nowing to their collaborative nature across different tasks. Yet, the\\nimplications of adversarial attacks on such models remain relatively\\nunexplored. In this paper, we conduct comprehensive adversarial security\\nresearch on the modular end-to-end autonomous driving model for the first time.\\nWe thoroughly consider the potential vulnerabilities in the model inference\\nprocess and design a universal attack scheme through module-wise noise\\ninjection. We conduct large-scale experiments on the full-stack autonomous\\ndriving model and demonstrate that our attack method outperforms previous\\nattack methods. We trust that our research will offer fresh insights into\\nensuring the safety and reliability of autonomous driving systems.\",\"PeriodicalId\":501301,\"journal\":{\"name\":\"arXiv - CS - Machine Learning\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-12\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Machine Learning\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.07706\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Machine Learning","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.07706","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

摘要

随着最近深度神经网络的突破,许多自动驾驶任务都表现出了不俗的性能。然而,深度学习模型容易受到对抗性攻击,给自动驾驶系统带来了巨大的安全风险。目前,端到端体系结构已成为自动驾驶的主要解决方案,因为它们具有跨不同任务的协作性质。然而,对抗性攻击对此类模型的影响仍然相对较少。本文首次对模块化端到端自动驾驶模型进行了全面的对抗性安全研究。我们深入考虑了模型推理过程中的潜在漏洞,并设计了一种通过模块噪声注入的通用攻击方案。我们在全栈自动驾驶模型上进行了大规模实验,证明我们的攻击方法优于之前的攻击方法。我们相信,我们的研究将为确保自动驾驶系统的安全性和可靠性提供新的见解。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
Attack End-to-End Autonomous Driving through Module-Wise Noise
With recent breakthroughs in deep neural networks, numerous tasks within autonomous driving have exhibited remarkable performance. However, deep learning models are susceptible to adversarial attacks, presenting significant security risks to autonomous driving systems. Presently, end-to-end architectures have emerged as the predominant solution for autonomous driving, owing to their collaborative nature across different tasks. Yet, the implications of adversarial attacks on such models remain relatively unexplored. In this paper, we conduct comprehensive adversarial security research on the modular end-to-end autonomous driving model for the first time. We thoroughly consider the potential vulnerabilities in the model inference process and design a universal attack scheme through module-wise noise injection. We conduct large-scale experiments on the full-stack autonomous driving model and demonstrate that our attack method outperforms previous attack methods. We trust that our research will offer fresh insights into ensuring the safety and reliability of autonomous driving systems.
求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信