Realistic adversarial attacks on various camera-based perception tasks of autonomous vehicles have been successfully demonstrated so far. However, only a few works considered attacks on traffic light detectors. This work shows how CNNs for traffic light detection can be attacked with printed patches. We propose a threat model, where each instance of a traffic light is attacked with a patch placed under it, and describe a training strategy. We demonstrate successful adversarial patch attacks in universal settings. Our experiments show realistic targeted red-to-green label-flipping attacks and attacks on pictogram classification. Finally, we perform a real-world evaluation with printed patches and demonstrate attacks in the lab settings with a mobile traffic light for construction sites and in a test area with stationary traffic lights. Our code is available at https://github.com/KASTEL-MobilityLab/attacks-on-traffic-light-detection.
@article{arxiv.2506.04823,
title = {Fool the Stoplight: Realistic Adversarial Patch Attacks on Traffic Light Detectors},
author = {Svetlana Pavlitska and Jamie Robb and Nikolai Polley and Melih Yazgan and J. Marius Zöllner},
journal= {arXiv preprint arXiv:2506.04823},
year = {2025}
}