MCN-SLAM: Multi-Agent Collaborative Neural SLAM with Hybrid Implicit Neural Scene Representation

2025-06-23

AI Summary

Existing neural implicit SLAM methods are limited to single-agent, small-scale scenes, and short sequences; multi-agent NeRF-SLAM frameworks struggle to satisfy communication bandwidth constraints, and no real-world multi-agent dataset provides both continuous-time ground-truth trajectories and high-fidelity 3D mesh ground truth. This paper proposes the first distributed multi-agent collaborative neural SLAM framework tailored for communication-constrained environments. It innovatively integrates a triplane-mesh joint implicit representation, an intra-to-inter loop closure detection mechanism, and online knowledge distillation–driven multi-subgraph fusion. The framework enables long-sequence dense mapping and high-precision localization in large-scale scenes. Experiments demonstrate superior performance over state-of-the-art methods in mapping accuracy, pose estimation, and communication efficiency. Additionally, we release DES—the first real-world multi-agent dense SLAM dataset featuring high-accuracy continuous-time trajectory and 3D mesh ground truth.

Technology Category

• Multiagent Systems: Multiagent Planning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Machine Learning: Distributed Machine Learning & Federated Learning

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Neural implicit scene representations have recently shown promising results in dense visual SLAM. However, existing implicit SLAM algorithms are constrained to single-agent scenarios, and fall difficulties in large-scale scenes and long sequences. Existing NeRF-based multi-agent SLAM frameworks cannot meet the constraints of communication bandwidth. To this end, we propose the first distributed multi-agent collaborative neural SLAM framework with hybrid scene representation, distributed camera tracking, intra-to-inter loop closure, and online distillation for multiple submap fusion. A novel triplane-grid joint scene representation method is proposed to improve scene reconstruction. A novel intra-to-inter loop closure method is designed to achieve local (single-agent) and global (multi-agent) consistency. We also design a novel online distillation method to fuse the information of different submaps to achieve global consistency. Furthermore, to the best of our knowledge, there is no real-world dataset for NeRF-based/GS-based SLAM that provides both continuous-time trajectories groundtruth and high-accuracy 3D meshes groundtruth. To this end, we propose the first real-world Dense slam (DES) dataset covering both single-agent and multi-agent scenarios, ranging from small rooms to large-scale outdoor scenes, with high-accuracy ground truth for both 3D mesh and continuous-time camera trajectory. This dataset can advance the development of the research in both SLAM, 3D reconstruction, and visual foundation model. Experiments on various datasets demonstrate the superiority of the proposed method in both mapping, tracking, and communication. The dataset and code will open-source on https://github.com/dtc111111/mcnslam.

Problem

• Extends neural SLAM to multi-agent collaborative scenarios
• Addresses large-scale scene and long sequence challenges
• Solves communication bandwidth constraints in multi-agent SLAM

Innovation

• Hybrid implicit neural scene representation
• Distributed multi-agent camera tracking
• Online distillation for submap fusion