DRACo-SLAM2: Distributed Robust Acoustic Communication-efficient SLAM for Imaging Sonar EquippedUnderwater Robot Teams with Object Graph Matching

2025-07-31

AI Summary

In underwater multi-robot collaborative SLAM, the absence of geometric priors and the low efficiency and poor robustness of inter-robot loop closure detection pose significant challenges. To address these, this paper proposes a distributed SLAM framework based on an Object Graph representation. Its key contributions are: (1) adopting a semantics-augmented Object Graph as a unified sonar map representation, eliminating reliance on geometric priors; (2) designing a lightweight Object Graph matching mechanism for efficient and robust inter-robot loop closure detection; and (3) introducing an incremental Group-wise Consistent Measurement maximization (GCM) algorithm to replace conventional Pose-Constraint Matching (PCM), thereby mitigating error propagation from shared nearby loop closures. Evaluated on both simulated and real-world underwater multibeam sonar datasets, the method achieves a 28.6% increase in loop closure detection rate, a 32.4% reduction in absolute trajectory error (ATE), and a 41% decrease in communication overhead—demonstrating strong performance in complex, texture-poor, and low-overlap underwater environments.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Graph-based Machine Learning
• Search and Optimization: Distributed Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

We present DRACo-SLAM2, a distributed SLAM framework for underwater robot teams equipped with multibeam imaging sonar. This framework improves upon the original DRACo-SLAM by introducing a novel representation of sonar maps as object graphs and utilizing object graph matching to achieve time-efficient inter-robot loop closure detection without relying on prior geometric information. To better-accommodate the needs and characteristics of underwater scan matching, we propose incremental Group-wise Consistent Measurement Set Maximization (GCM), a modification of Pairwise Consistent Measurement Set Maximization (PCM), which effectively handles scenarios where nearby inter-robot loop closures share similar registration errors. The proposed approach is validated through extensive comparative analyses on simulated and real-world datasets.

Problem

• Distributed SLAM for underwater robot teams with sonar
• Efficient loop closure detection using object graph matching
• Improved scan matching for shared registration errors

Innovation

• Object graph matching for loop closure
• Incremental Group-wise Consistent Measurement Maximization
• Distributed SLAM for underwater robot teams