CoDynTrust: Robust Asynchronous Collaborative Perception via Dynamic Feature Trust Modulus

2025-02-12

AI Summary

To address temporal misalignment in multi-vehicle collaborative perception—caused by communication latency, clock skew, and asynchronous sampling—this paper proposes an uncertainty-driven dynamic feature credibility modulation framework. The method quantifies both aleatoric and epistemic uncertainties for each vehicle’s features, introduces the Dynamic Feature Trust Modulus (DFTM) to adaptively suppress or retain features based on their reliability, and explicitly propagates uncertainty to downstream planning and control modules. It further incorporates dynamic weighted modulation, multi-scale asynchronous fusion, and an end-to-end collaborative detection architecture. Evaluated on multiple benchmark datasets, the approach significantly mitigates performance degradation under asynchrony and achieves state-of-the-art detection accuracy in asynchronous multi-vehicle settings. The implementation is publicly available.

Technology Category

• Computer Vision: Multi-modal Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Collaborative perception, fusing information from multiple agents, can extend perception range so as to improve perception performance. However, temporal asynchrony in real-world environments, caused by communication delays, clock misalignment, or sampling configuration differences, can lead to information mismatches. If this is not well handled, then the collaborative performance is patchy, and what's worse safety accidents may occur. To tackle this challenge, we propose CoDynTrust, an uncertainty-encoded asynchronous fusion perception framework that is robust to the information mismatches caused by temporal asynchrony. CoDynTrust generates dynamic feature trust modulus (DFTM) for each region of interest by modeling aleatoric and epistemic uncertainty as well as selectively suppressing or retaining single-vehicle features, thereby mitigating information mismatches. We then design a multi-scale fusion module to handle multi-scale feature maps processed by DFTM. Compared to existing works that also consider asynchronous collaborative perception, CoDynTrust combats various low-quality information in temporally asynchronous scenarios and allows uncertainty to be propagated to downstream tasks such as planning and control. Experimental results demonstrate that CoDynTrust significantly reduces performance degradation caused by temporal asynchrony across multiple datasets, achieving state-of-the-art detection performance even with temporal asynchrony. The code is available at https://github.com/CrazyShout/CoDynTrust.

Problem

• Address temporal asynchrony in collaborative perception
• Mitigate information mismatches via dynamic feature trust
• Enhance perception performance in asynchronous environments

Innovation

• Dynamic Feature Trust Modulus
• Uncertainty-encoded fusion framework
• Multi-scale feature fusion module