TY - JOUR
T1 - Coupling Structural Descriptors with a Novel Semantic Graph Matching Approach for LiDAR Loop Detection
AU - Tang, Yujie
AU - Wang, Meiling
AU - Zuo, Sibo
AU - Yang, Chengxi
AU - Jiang, Jinhao
AU - Chen, Xieyuanli
AU - Yue, Yufeng
N1 - Publisher Copyright:
© 2004-2012 IEEE.
PY - 2026
Y1 - 2026
N2 - Outdoor loop closure detection is essential for correcting odometry drift and constructing a globally consistent map. Semantic-graph-based approaches effectively model object-level topology and achieve strong loop closure performance; however, their effectiveness degrades in background-dominated scenes with few distinctive objects, and establishing accurate injective node correspondences remains challenging. In contrast, structural descriptor methods, though offering stronger environmental generality through spatial-distribution modeling, remain susceptible to LiDAR noise and the discriminative power of point-level features. These limitations motivate the need for a more robust method that combines adaptability with enhanced descriptive power. We propose a novel loop-closure detection framework, SAGE, that integrates highly adaptable point-cloud shape-distribution features and generally reliable semantic graph topology, adaptively combining their similarity measures to improve detection performance. Specifically, we design a semantic graph matching module with dual constraints, local graph feature consistency and global spatial consistency, to achieve more accurate injective node correspondences. In addition, we extract point-cloud shape-distribution features and introduce a fusion mechanism that integrates them with the semantic graph module, assessing reliability and adaptively weighting their contributions. Extensive loop closure detection and pose estimation experiments on various datasets demonstrate that SAGE achieves superior performance over strong baselines. We provide the code at https://github.com/SAGE-11/SAGE Note to Practitioners - This paper focuses on improving the accuracy and robustness of LiDAR loop closure detection in outdoor road environments, which is a critical component for long-term robot navigation, and mapping. Mainstream semantic-graph-based methods typically require a number of nodes in the semantic graph and accurate estimates of object centers; otherwise, the correct loop closure can be difficult to identify, significantly limiting the environments applicable. To address this limitation, we propose SAGE, a framework that integrates semantic graph matching with more widely applicable structural descriptors through an effective fusion strategy. This design enhances loop closure detection in both accuracy and robustness. Future work may extend our method beyond road environments to a wider range of scenarios, including forests, valleys, and other natural environments.
AB - Outdoor loop closure detection is essential for correcting odometry drift and constructing a globally consistent map. Semantic-graph-based approaches effectively model object-level topology and achieve strong loop closure performance; however, their effectiveness degrades in background-dominated scenes with few distinctive objects, and establishing accurate injective node correspondences remains challenging. In contrast, structural descriptor methods, though offering stronger environmental generality through spatial-distribution modeling, remain susceptible to LiDAR noise and the discriminative power of point-level features. These limitations motivate the need for a more robust method that combines adaptability with enhanced descriptive power. We propose a novel loop-closure detection framework, SAGE, that integrates highly adaptable point-cloud shape-distribution features and generally reliable semantic graph topology, adaptively combining their similarity measures to improve detection performance. Specifically, we design a semantic graph matching module with dual constraints, local graph feature consistency and global spatial consistency, to achieve more accurate injective node correspondences. In addition, we extract point-cloud shape-distribution features and introduce a fusion mechanism that integrates them with the semantic graph module, assessing reliability and adaptively weighting their contributions. Extensive loop closure detection and pose estimation experiments on various datasets demonstrate that SAGE achieves superior performance over strong baselines. We provide the code at https://github.com/SAGE-11/SAGE Note to Practitioners - This paper focuses on improving the accuracy and robustness of LiDAR loop closure detection in outdoor road environments, which is a critical component for long-term robot navigation, and mapping. Mainstream semantic-graph-based methods typically require a number of nodes in the semantic graph and accurate estimates of object centers; otherwise, the correct loop closure can be difficult to identify, significantly limiting the environments applicable. To address this limitation, we propose SAGE, a framework that integrates semantic graph matching with more widely applicable structural descriptors through an effective fusion strategy. This design enhances loop closure detection in both accuracy and robustness. Future work may extend our method beyond road environments to a wider range of scenarios, including forests, valleys, and other natural environments.
KW - Loop closure detection
KW - semantic graph matching
KW - structural descriptor
UR - https://www.scopus.com/pages/publications/105034202167
U2 - 10.1109/TASE.2026.3676786
DO - 10.1109/TASE.2026.3676786
M3 - Article
AN - SCOPUS:105034202167
SN - 1545-5955
VL - 23
SP - 7013
EP - 7024
JO - IEEE Transactions on Automation Science and Engineering
JF - IEEE Transactions on Automation Science and Engineering
ER -