SlideSLAM: Sparse, Lightweight, Decentralized Metric-Semantic SLAM for Multi-Robot Navigation

2024-06-25

arXiv.org

AI Summary

To address collaborative mapping and localization for multi-robot systems under GPS-denied conditions, this paper proposes a lightweight, sparse, and decentralized metric-semantic SLAM framework enabling real-time joint mapping by heterogeneous aerial-ground robots in complex environments—including indoor spaces, urban outdoors, and forested areas. Our method integrates semantic-aware perception with distributed optimization: (1) a novel semantics-driven cross-robot loop closure mechanism achieves viewpoint-invariant, object-level place recognition; (2) an instance-segmentation-based front-end coupled with a custom graph-optimization back-end jointly fuses semantic-enhanced loop closures and distributed communication-state tracking; (3) opportunistic communication and resource-adaptive coordination are natively supported. Experiments demonstrate significant improvements in inter-robot localization accuracy and object-level map consistency, while reducing computational load, memory footprint, and communication overhead. The framework is open-sourced and supports both single- and multi-robot deployments.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Multiagent Systems: Coordination and Collaboration

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

This paper develops a real-time decentralized metric-semantic Simultaneous Localization and Mapping (SLAM) algorithm framework that enables a heterogeneous robot team to collaboratively construct object-based metric-semantic maps of real-world environments featuring indoor, urban, and forests without relying on GPS. The framework integrates a data-driven front-end for instance segmentation from either RGBD cameras or LiDARs and a custom back-end for optimizing robot trajectories and object landmarks in the map. To allow multiple robots to merge their information, we design semantics-driven place recognition algorithms that leverage the informativeness and viewpoint invariance of the object-level metric-semantic map for inter-robot loop closure detection. A communication module is designed to track each robot's observations and those of other robots whenever communication links are available. Our framework enables real-time decentralized operations onboard robots, allowing them to leverage communication opportunistically. We integrate the proposed framework with the autonomous navigation and exploration systems of three types of aerial and ground robots, conducting extensive experiments in a variety of indoor and outdoor environments. These experiments demonstrate its accuracy in inter-robot localization and object mapping, along with its moderate demands on computation, storage, and communication resources. The framework is open-sourced and is suitable for both single-agent and multi-robot metric-semantic SLAM applications. The project website and code can be found at https://xurobotics.github.io/slideslam/ and https://github.com/XuRobotics/SLIDE_SLAM, respectively.

Problem

• Multi-robot systems
• Indoor/Outdoor mapping
• GPS-denied environments

Innovation

• Multi-Robot SLAM
• Decentralized Control
• Sensor Integration