Quantcast

Central OC Times

Saturday, March 8, 2025

Study reveals potential risks in self-driving car sign recognition systems

Webp ip7n8zy0568cekq7bm4fbrj37or5

Chancellor Emeritus Michael V. Drake, M.D. | Official website

Chancellor Emeritus Michael V. Drake, M.D. | Official website

Researchers at the University of California, Irvine have uncovered a significant vulnerability in self-driving vehicle technology. Their study reveals that multicolored stickers applied to roadside signs can confuse autonomous vehicles, leading to potentially dangerous situations. This research was presented at the Network and Distributed System Security Symposium in San Diego.

The team from UC Irvine's Donald Bren School of Information & Computer Sciences demonstrated that these "sticker attacks" could make traffic signs undetectable to artificial intelligence algorithms in some driverless cars. Such attacks could result in vehicles ignoring road commands or executing unintended actions like emergency braking or speeding.

Alfred Chen, co-author and assistant professor of computer science at UC Irvine, highlighted the growing prevalence of autonomous technology: “Waymo has been delivering more than 150,000 autonomous rides per week, and there are millions of Autopilot-equipped Tesla vehicles on the road." He emphasized the importance of security as vulnerabilities could pose life-threatening hazards.

Ningfei Wang, lead author and research scientist at Meta who conducted this work during his Ph.D. studies at UC Irvine, explained their method involved using stickers with swirling designs to disrupt AI algorithms responsible for traffic sign recognition. According to Wang, these stickers can be easily produced using accessible tools like Python and image processing libraries.

Wang also noted an interesting discovery regarding spatial memorization design common in commercial traffic sign recognition systems. While it complicates disappearing attacks (removing signs from view), it simplifies spoofing fake stop signs more than anticipated.

Chen pointed out that their research is pioneering in evaluating this security threat with commercially available vehicles. “Academics have studied driverless vehicle security for years... Our study fills this critical gap,” he said.

He further stressed the need for continued exploration into these vulnerabilities: “We believe this work should only be the beginning... action is needed to ensure safety on our streets and highways.”

The project included contributions from former UC Irvine graduate students Takami Sato and Yunpeng Luo; current graduate student Shaoyuan Xie; and Kaidi Xu from Drexel University. The National Science Foundation and the U.S. Department of Transportation’s CARMEN+ University Transportation Center supported this research.

UC Irvine's Brilliant Future campaign aims to bolster university excellence through alumni engagement and philanthropic investment. Founded in 1965, UC Irvine is recognized for its academic achievements and economic contributions both locally and statewide.

For more information about UC Irvine or media inquiries, visit news resources provided by the university.

ORGANIZATIONS IN THIS STORY

!RECEIVE ALERTS

The next time we write about any of these orgs, we’ll email you a link to the story. You may edit your settings or unsubscribe at any time.
Sign-up

DONATE

Help support the Metric Media Foundation's mission to restore community based news.
Donate