A Survey on Collaborative SLAM with 3D Gaussian Splatting

2025-10-27

AI Summary

This paper addresses core challenges—consistency maintenance, heterogeneous data fusion, and communication efficiency—in multi-robot collaborative SLAM when integrating 3D Gaussian Splatting (3DGS). It presents the first systematic survey of 3DGS-based collaborative SLAM approaches. To structure the analysis, we propose a three-dimensional framework: “multi-agent consistency alignment—semantic distillation—lightweight communication fusion,” clarifying co-optimization pathways for high-fidelity mapping and real-time rendering under distributed and centralized architectures. We comprehensively review prevailing datasets, evaluation metrics, and empirical performance bounds. Key open problems are identified, including long-term mapping, semantic grounding, multimodal robustness, and Sim2Real transferability. The work delivers the first technical roadmap and reproducible benchmark suite for this emerging interdisciplinary domain, enabling rigorous comparative evaluation and guiding future research directions.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Multiagent Systems: Coordination and Collaboration
• Computer Vision: 3D Computer Vision

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

This survey comprehensively reviews the evolving field of multi-robot collaborative Simultaneous Localization and Mapping (SLAM) using 3D Gaussian Splatting (3DGS). As an explicit scene representation, 3DGS has enabled unprecedented real-time, high-fidelity render- ing, ideal for robotics. However, its use in multi-robot systems introduces significant challenges in maintaining global consistency, managing communication, and fusing data from heterogeneous sources. We systematically categorize approaches by their architecture-centralized, distributed- and analyze core components like multi-agent consistency and alignment, communication- efficient, Gaussian representation, semantic distillation, fusion and pose optimization, and real- time scalability. In addition, a summary of critical datasets and evaluation metrics is provided to contextualize performance. Finally, we identify key open challenges and chart future research directions, including lifelong mapping, semantic association and mapping, multi-model for robustness, and bridging the Sim2Real gap.

Problem

• Surveying multi-robot collaborative SLAM using 3D Gaussian Splatting techniques
• Addressing global consistency and data fusion challenges in heterogeneous systems
• Analyzing communication efficiency and scalability for real-time robotic applications

Innovation

• Multi-robot collaborative SLAM using 3D Gaussian Splatting
• Systematic categorization of centralized and distributed architectures
• Addressing global consistency and communication efficiency challenges