Uncertainty Quantification for Collaborative Object Detection Under Adversarial Attacks

2025-02-04

AI Summary

Collaborative object detection (COD) in vehicle-infrastructure cooperative systems suffers from insufficient robustness and inadequate uncertainty quantification under adversarial attacks. Method: This paper proposes TUQCP, a Trustworthy Uncertainty Quantification framework for COD, which—novelty—introduces conformal prediction calibration into collaborative perception. TUQCP integrates adversarial training with a learnable uncertainty estimation module, supporting both early- and mid-fusion collaboration as well as single-agent models, and enabling reliable confidence calibration even under unknown attack patterns. Contribution/Results: Evaluated on the V2X-Sim simulation dataset, TUQCP improves adversarial robustness by boosting detection accuracy by 80.41% under attacks. It significantly enhances system robustness, interpretability, and decision trustworthiness, establishing a verifiable uncertainty assurance mechanism for dynamic autonomous driving scenarios.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Adversarial Attacks & Robustness
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Collaborative Object Detection (COD) and collaborative perception can integrate data or features from various entities, and improve object detection accuracy compared with individual perception. However, adversarial attacks pose a potential threat to the deep learning COD models, and introduce high output uncertainty. With unknown attack models, it becomes even more challenging to improve COD resiliency and quantify the output uncertainty for highly dynamic perception scenes such as autonomous vehicles. In this study, we propose the Trusted Uncertainty Quantification in Collaborative Perception framework (TUQCP). TUQCP leverages both adversarial training and uncertainty quantification techniques to enhance the adversarial robustness of existing COD models. More specifically, TUQCP first adds perturbations to the shared information of randomly selected agents during object detection collaboration by adversarial training. TUQCP then alleviates the impacts of adversarial attacks by providing output uncertainty estimation through learning-based module and uncertainty calibration through conformal prediction. Our framework works for early and intermediate collaboration COD models and single-agent object detection models. We evaluate TUQCP on V2X-Sim, a comprehensive collaborative perception dataset for autonomous driving, and demonstrate a 80.41% improvement in object detection accuracy compared to the baselines under the same adversarial attacks. TUQCP demonstrates the importance of uncertainty quantification to COD under adversarial attacks.

Problem

• Quantify uncertainty in collaborative object detection
• Enhance adversarial robustness in COD models
• Improve accuracy under adversarial attacks in autonomous vehicles

Innovation

• Adversarial training enhances robustness
• Uncertainty quantification improves accuracy
• Conformal prediction calibrates uncertainty