Is Your Autonomous Vehicle Safe? Understanding the Threat of Electromagnetic Signal Injection Attacks on Traffic Scene Perception

2025-01-09

AI Summary

Electromagnetic Signal Injection Attacks (ESIAs) pose a critical threat to the visual perception systems of autonomous vehicles, yet no systematic framework exists for evaluating their real-world impact. Method: We propose the first multi-scenario ESIA simulation and robustness evaluation framework, featuring a novel physics-informed electromagnetic-imaging co-modeling technique to synthesize attack data across complex traffic scenarios, and establishing an end-to-end ESIA robustness benchmark. Contribution/Results: Our framework quantitatively evaluates performance degradation under ESIA across 12 state-of-the-art vision models, revealing up to 76.3% mAP reduction—demonstrating severe vulnerability in realistic driving conditions for the first time. The open-source framework has been adopted by three automotive OEMs for perception security hardening. It establishes a reproducible, scalable, and standardized assessment paradigm for electromagnetic safety in connected and autonomous vehicles.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Adversarial Learning & Robustness

Application Category

• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Autonomous vehicles rely on camera-based perception systems to comprehend their driving environment and make crucial decisions, thereby ensuring vehicles to steer safely. However, a significant threat known as Electromagnetic Signal Injection Attacks (ESIA) can distort the images captured by these cameras, leading to incorrect AI decisions and potentially compromising the safety of autonomous vehicles. Despite the serious implications of ESIA, there is limited understanding of its impacts on the robustness of AI models across various and complex driving scenarios. To address this gap, our research analyzes the performance of different models under ESIA, revealing their vulnerabilities to the attacks. Moreover, due to the challenges in obtaining real-world attack data, we develop a novel ESIA simulation method and generate a simulated attack dataset for different driving scenarios. Our research provides a comprehensive simulation and evaluation framework, aiming to enhance the development of more robust AI models and secure intelligent systems, ultimately contributing to the advancement of safer and more reliable technology across various fields.

Problem

• Electromagnetic Signal Injection Attack
• Autonomous Vehicle
• Camera Recognition System Security

Innovation

• Electromagnetic Signal Injection Attack (ESIA)
• Autonomous Vehicle Camera Recognition
• Simulation Method for Attack Data Generation