PaniCar: Securing the Perception of Advanced Driving Assistance Systems Against Emergency Vehicle Lighting

2025-05-08

AI Summary

This paper identifies and addresses PaniCar—a critical digital-domain vulnerability in ADAS object detectors wherein intense glare from emergency vehicle lights causes severe confidence collapse, jeopardizing autonomous driving safety. We present Caracetamol, a lightweight, real-time, and robust detection enhancement framework that improves glare resilience without image restoration. Caracetamol jointly employs confidence calibration and feature-domain adversarial suppression to mitigate glare-induced degradation. It is plug-and-play compatible with mainstream detectors (e.g., YOLOv3, Faster R-CNN) and deployable on embedded ADAS platforms. Experiments demonstrate that under strong glare, Caracetamol increases average detection confidence by 0.20, raises the confidence lower bound by 0.33, and reduces confidence fluctuation range by 0.33, while sustaining real-time inference at 30–50 FPS.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Application Domains: Mobility, Driving & Flight
• Machine Learning: Adversarial Learning & Robustness

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Sustainability and carbon-aware systems for Web, mobile, and WoT
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Security and Privacy: Large-scale security measurements

Abstract

The safety of autonomous cars has come under scrutiny in recent years, especially after 16 documented incidents involving Teslas (with autopilot engaged) crashing into parked emergency vehicles (police cars, ambulances, and firetrucks). While previous studies have revealed that strong light sources often introduce flare artifacts in the captured image, which degrade the image quality, the impact of flare on object detection performance remains unclear. In this research, we unveil PaniCar, a digital phenomenon that causes an object detector's confidence score to fluctuate below detection thresholds when exposed to activated emergency vehicle lighting. This vulnerability poses a significant safety risk, and can cause autonomous vehicles to fail to detect objects near emergency vehicles. In addition, this vulnerability could be exploited by adversaries to compromise the security of advanced driving assistance systems (ADASs). We assess seven commercial ADASs (Tesla Model 3,"manufacturer C", HP, Pelsee, AZDOME, Imagebon, Rexing), four object detectors (YOLO, SSD, RetinaNet, Faster R-CNN), and 14 patterns of emergency vehicle lighting to understand the influence of various technical and environmental factors. We also evaluate four SOTA flare removal methods and show that their performance and latency are insufficient for real-time driving constraints. To mitigate this risk, we propose Caracetamol, a robust framework designed to enhance the resilience of object detectors against the effects of activated emergency vehicle lighting. Our evaluation shows that on YOLOv3 and Faster RCNN, Caracetamol improves the models' average confidence of car detection by 0.20, the lower confidence bound by 0.33, and reduces the fluctuation range by 0.33. In addition, Caracetamol is capable of processing frames at a rate of between 30-50 FPS, enabling real-time ADAS car detection.

Problem

• Impact of emergency vehicle lighting on autonomous car object detection
• Vulnerability of ADASs to adversarial exploitation via lighting patterns
• Insufficiency of current flare removal methods for real-time driving

Innovation

• Proposes Caracetamol to enhance detector resilience
• Evaluates flare removal methods for real-time driving
• Assesses impact of emergency lighting on object detection