Simultaneous Localization and Mapping (SLAM) is crucial for achieving autonomous construction. This paper presents a 3D LiDAR-based SLAM system specifically integrated and optimized for construction robots operating in large-scale public building construction sites. Rather than introducing new algorithms, the proposed system integrates and optimizes existing modules within a unified system framework, including two-stage ground segmentation, Fast Euclidean Clustering (FEC) for dynamic-object suppression, two-step point cloud registration for odometry, and Scan Context++ for global pose optimization. The system design leverages the ground characteristics of building floors and addresses dynamic, unstructured, and repetitive spatial structures commonly found in building construction site environments. Both simulation and real-world experiments demonstrate that the proposed system achieves higher accuracy and robustness than benchmark SLAM methods while maintaining real-time performance on embedded hardware, confirming its effectiveness for autonomous construction applications. • Integrated LiDAR SLAM system tailored for large-scale public building construction sites. • Combines ground segmentation, clustering, registration, and loop closure into one framework. • Verified through both Gazebo-based simulation and real construction site experiments. • Provides a reliable localization and mapping foundation for autonomous construction robots.
Feng et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: