English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In autonomous driving (AD), accurate perception is indispensable to achievingsafe and secure driving. Due to its safety-criticality, the security of ADperception has been widely studied. Among different attacks on AD perception,the physical adversarial object evasion attacks are especially severe. However,we find that all existing literature only evaluates their attack effect at thetargeted AI component level but not at the system level, i.e., with the entiresystem semantics and context such as the full AD pipeline. Thereby, this raisesa critical research question: can these existing researches effectively achievesystem-level attack effects (e.g., traffic rule violations) in the real-worldAD context? In this work, we conduct the first measurement study on whether andhow effectively the existing designs can lead to system-level effects,especially for the STOP sign-evasion attacks due to their popularity andseverity. Our evaluation results show that all the representative prior workscannot achieve any system-level effects. We observe two design limitations inthe prior works: 1) physical model-inconsistent object size distribution inpixel sampling and 2) lack of vehicle plant model and AD system modelconsideration. Then, we propose SysAdv, a novel system-driven attack design inthe AD context and our evaluation results show that the system-level effectscan be significantly improved, i.e., the violation rate increases by around70%.

Does Physical Adversarial Example Really Matter to Autonomous Driving? Towards System-Level Effect of Adversarial Object Evasion Attack