Cross-Modal Phantom: Coordinated Camera-LiDAR Spoofing Against Multi-Sensor Fusion in Autonomous Vehicles

TL;DR

This paper demonstrates a novel coordinated spoofing attack on autonomous vehicle perception systems that can deceive multi-sensor fusion by fabricating consistent false signals across camera and LiDAR sensors.

Contribution

It introduces a simulated cross-modal spoofing method that exposes a critical vulnerability in multi-sensor fusion for autonomous vehicles, achieving high success rates in deception.

Findings

85.5% success rate in deceiving perception models

Simulated sensor-level spoofing mimics physical IR and LiDAR attacks

Reveals a critical vulnerability in multi-sensor fusion systems

Abstract

Autonomous Vehicles (AVs) increasingly depend on Multi-Sensor Fusion (MSF) to combine complementary modalities such as cameras and LiDAR for robust perception. While this redundancy is intended to safeguard against single-sensor failures, the fusion process itself introduces a subtle and underexplored vulnerability. In this work, we investigate whether an attacker can bypass MSF's redundancy by fabricating cross-sensor consistency, making multiple sensors agree on the same false object. We design a coordinated, data-level (early-fusion) attack that emulates the outcome of two synchronized physical spoofing sources: an infrared (IR) projection that induces a false camera detection and a LiDAR signal injection that produces a matching 3D point cluster. Rather than implementing the physical attack hardware, we simulate its sensor-level outcomes by inserting perspective-aware image patches and synthetic LiDAR point clusters aligned in 3D space. This approach preserves the perceptual effects that real IR and IEMI-based spoofing would create at the sensor output. Using 400 KITTI scenes, our large-scale evaluation shows that the coordinated spoofing deceives a state-of-the-art perception model with an 85.5% successful attack rate. These findings provide the first quantitative evidence that malicious cross-modal consistency can compromise MSF-based perception, revealing a critical vulnerability in the core data-fusion logic of modern autonomous vehicle systems.