Search papers, labs, and topics across Lattice.
This paper presents SGM-SLAM, a novel data-efficient distributed SLAM framework that leverages scene graph matching for inter-robot measurement constraints, utilizing only object labels and centroids instead of traditional feature-level matching. By constructing a scene graph from fused RGB-LiDAR point clouds, the framework enhances the accuracy of robot localization and mapping while minimizing data exchange. Experimental results from simulations and real-world datasets demonstrate significant improvements in efficiency and effectiveness compared to existing methods.
Scene graph matching enables robots to achieve accurate localization and mapping with minimal data exchange, revolutionizing distributed SLAM efficiency.
We introduce a data-efficient distributed Simultaneous Localization and Mapping (SLAM) framework designed for a team of robots equipped with LiDAR, cameras, and inertial sensors. Our framework uses scene graph matching to identify inter-robot measurement constraints. Unlike prior approaches that rely on feature-level matching, our framework is the first to perform scene graph matching using only object labels and centroids. Our approach constructs a scene graph by using fused RGB-LiDAR point clouds to generate both a semantically segmented point cloud layer, and a layer of discrete bounded objects, to accompany estimated robot trajectories. Scene graph matching is performed collaboratively through exchanging and matching object data with neighboring robots. To maximize communication efficiency, we utilize a multi-step data exchange and optimization process. We demonstrate the effectiveness and efficiency of our approach using both simulation and real-world datasets collected by legged robots in indoor and outdoor environments.
