CP-Guard+: A New Paradigm for Malicious Agent Detection and Defense in Collaborative Perception

TL;DR

This paper introduces CP-Guard+, a novel defense framework for collaborative perception systems that detects malicious agents at the feature level, backed by a new dataset and extensive experiments demonstrating its effectiveness.

Contribution

It proposes a new paradigm for malicious agent detection, introduces the CP-GuardBench dataset, and develops CP-Guard+ with a dual-centered contrastive loss for robust defense.

Findings

CP-Guard+ outperforms existing methods in detection accuracy.

The new paradigm reduces computational overhead in malicious detection.

Extensive experiments validate the effectiveness of CP-Guard+ on multiple datasets.

Abstract

Collaborative perception (CP) is a promising method for safe connected and autonomous driving, which enables multiple vehicles to share sensing information to enhance perception performance. However, compared with single-vehicle perception, the openness of a CP system makes it more vulnerable to malicious attacks that can inject malicious information to mislead the perception of an ego vehicle, resulting in severe risks for safe driving. To mitigate such vulnerability, we first propose a new paradigm for malicious agent detection that effectively identifies malicious agents at the feature level without requiring verification of final perception results, significantly reducing computational overhead. Building on this paradigm, we introduce CP-GuardBench, the first comprehensive dataset provided to train and evaluate various malicious agent detection methods for CP systems. Furthermore, we develop a robust defense method called CP-Guard+, which enhances the margin between the representations of benign and malicious features through a carefully designed Dual-Centered Contrastive Loss (DCCLoss). Finally, we conduct extensive experiments on both CP-GuardBench and V2X-Sim, and demonstrate the superiority of CP-Guard+.