LGCRCVMar 4, 2024

COMMIT: Certifying Robustness of Multi-Sensor Fusion Systems against Semantic Attacks

arXiv:2403.02329v14 citationsh-index: 15AAAI
Originality Highly original
AI Analysis

This addresses safety-critical robustness certification for autonomous vehicle perception systems, representing a foundational advance rather than an incremental improvement.

The paper tackles the problem of certifying robustness of multi-sensor fusion systems in autonomous vehicles against semantic attacks like rotation and shifting, proposing the first certification framework (COMMIT) that shows certification for MSF models can be up to 48.39% higher than for single-modal models.

Multi-sensor fusion systems (MSFs) play a vital role as the perception module in modern autonomous vehicles (AVs). Therefore, ensuring their robustness against common and realistic adversarial semantic transformations, such as rotation and shifting in the physical world, is crucial for the safety of AVs. While empirical evidence suggests that MSFs exhibit improved robustness compared to single-modal models, they are still vulnerable to adversarial semantic transformations. Despite the proposal of empirical defenses, several works show that these defenses can be attacked again by new adaptive attacks. So far, there is no certified defense proposed for MSFs. In this work, we propose the first robustness certification framework COMMIT certify robustness of multi-sensor fusion systems against semantic attacks. In particular, we propose a practical anisotropic noise mechanism that leverages randomized smoothing with multi-modal data and performs a grid-based splitting method to characterize complex semantic transformations. We also propose efficient algorithms to compute the certification in terms of object detection accuracy and IoU for large-scale MSF models. Empirically, we evaluate the efficacy of COMMIT in different settings and provide a comprehensive benchmark of certified robustness for different MSF models using the CARLA simulation platform. We show that the certification for MSF models is at most 48.39% higher than that of single-modal models, which validates the advantages of MSF models. We believe our certification framework and benchmark will contribute an important step towards certifiably robust AVs in practice.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes