Robust Collaborative Perception without External Localization and Clock Devices

A consistent spatial-temporal coordination across multiple agents is fundamental for collaborative perception, which seeks to improve perception abilities through information exchange among agents. To achieve this spatial-temporal alignment, traditional methods depend on external devices to provide localization and clock signals. However, hardware-generated signals could be vulnerable to noise and potentially malicious attack, jeopardizing the precision of spatial-temporal alignment. Rather than relying on external hardwares, this work proposes a novel approach: aligning by recognizing the inherent geometric patterns within the perceptual data of various agents. Following this spirit, we propose a robust collaborative perception system that operates independently of external localization and clock devices. The key module of our system, FreeAlign, constructs a salient object graph for each agent based on its detected boxes and uses a graph neural network to identify common subgraphs between agents, leading to accurate relative pose and time. We validate FreeAlign on both real-world and simulated datasets. The results show that, the FreeAlign empowered robust collaborative perception system perform comparably to systems relying on precise localization and clock devices. ${\mathbf{Code}}$ will be released.

Authors

• Zixing Lei
• Zhenyang Ni
• Ruize Han
• Shuo Tang
• Chen Feng 0002
• Siheng Chen
• Yanfeng Wang 0001

Keywords

• Location awareness
• Performance evaluation
• Accuracy
• Noise
• Collaboration
• Pattern recognition
• Synchronization
• Local Devices
• External Devices
• External Clock
• Simulated Datasets
• Real-world Datasets
• Multiple Agents
• Graph Neural Networks
• Malicious Attacks
• Collaborative System
• Geometric Patterns
• Salient Object
• Relative Pose
• Feature Maps
• Detection Performance
• Object Detection
• Visual Features
• Spatial Domain
• Geometric Structure
• Presence Of Noise
• Edge Features
• Graph Matching
• Pair Of Nodes
• Temporal Domain
• Performance In The Presence
• Alignment Accuracy
• 3D Object Detection
• Temporal Alignment
• Global Localization