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 3D 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 opportunistically leverage communication. 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 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 available as a modular stack for object-level metric-semantic SLAM, suitable for both single-agent and multi-robot scenarios.
Algorithm: We develop a real-time decentralized metric-semantic SLAM framework that supports various types of aerial and ground robots, equipped with LiDAR or RGBD sensors. The framework features: (a) a semantics-driven place recognition algorithm that utilizes sparse and lightweight object-level semantic maps for loop closure and map merging, and (b) a decentralized multi-robot collaboration module that facilitates information sharing, even under intermittent communication conditions.
System integration: We integrate and deploy the proposed framework on a heterogeneous team of robots, demonstrating its capacity to enable semantics-in-the-loop autonomous navigation and exploration in various indoor and outdoor environments. The system operates in real-time, utilizing limited onboard computation and memory resources.
Experiments: We conduct extensive real-world experiments and provide thorough empirical results and analysis that highlight the efficiency, accuracy, and robustness of our system.
@article{liu2024slideslam,
title={SlideSLAM: Sparse, Lightweight, Decentralized Metric-Semantic SLAM for Multi-Robot Navigation},
author={Liu, Xu and Lei, Jiuzhou and Prabhu, Ankit and Tao, Yuezhan and Spasojevic, Igor and Chaudhari, Pratik and Atanasov, Nikolay and Kumar, Vijay},
journal={arXiv preprint arXiv:2406.17249},
year={2024}
}