AI & ML Breaks Assumption

MirrorDrift demonstrates a successful SLAM-targeted attack on production-grade 'secure' LiDARs using simple actuated mirrors rather than complex signal injection.

arXiv · March 13, 2026 · 2603.11364

Rokuto Nagata, Kenji Koide, Kazuma Ikeda, Ozora Sako, Shion Horie, Kentaro Yoshioka

Why it matters

It challenges the assumption that modern LiDAR defense mechanisms (like timing obfuscation) are sufficient to ensure geometric consistency. It reveals a fundamental physical vulnerability in scan-matching algorithms that practitioners in autonomous driving must address.

From the abstract

LiDAR SLAM provides high-accuracy localization but is fragile to point-cloud corruption because scan matching assumes geometric consistency. Prior physical attacks on LiDAR SLAM largely rely on LiDAR spoofing via external signal injection, which requires sensor-specific timing knowledge and is increasingly mitigated by modern defense mechanisms such as timing obfuscation and injection rejection. In this work, we show that specular reflection offers an injection-free alternative and demonstrate a