UniAda: Universal Adaptive Multi-objective Adversarial Attack for End-to-End Autonomous Driving Systems
UniAda:面向端到端自动驾驶系统的通用自适应多目标对抗攻击
机构 * Department of Computer Science, City University of Hong Kong(城市大学计算机科学系) ; School of Cyber Science and Technology, Shenzhen Campus of Sun Yat-sen University(中山大学深圳校区信息科学与技术学院)
专题命中 端到端驾驶 :autonomous driving(title,abstract)
AI总结 UniAda提出一种多目标白盒攻击方法,通过自适应加权方案优化,同时影响转向和速度控制,验证结果显示其在转向和速度偏差上优于现有方法。
Comments Published at IEEE Transactions on Reliability journal (2023)