Research Article
Cluster-based Accurate Localization in Noisy and Sparse Unmanned Sensing Systems
@INPROCEEDINGS{10.4108/eai.18-12-2025.2365296, author={Bingjie Han and Yingxu Lai and Haodi Ping}, title={Cluster-based Accurate Localization in Noisy and Sparse Unmanned Sensing Systems}, proceedings={Proceedings of the 13th International Conference on Identification, Information and Knowledge in the Internet of Things, IIKI 2025, 18-21 December 2025, Chengdu, China}, publisher={EAI}, proceedings_a={IIKI}, year={2026}, month={6}, keywords={ranging-based localization noisy and sparse sensing unmanned system}, doi={10.4108/eai.18-12-2025.2365296} }- Bingjie Han
Yingxu Lai
Haodi Ping
Year: 2026
Cluster-based Accurate Localization in Noisy and Sparse Unmanned Sensing Systems
IIKI
EAI
DOI: 10.4108/eai.18-12-2025.2365296
Abstract
In unmanned sensing systems, ranging-based localization over noisy and sparse measurements often converges to stable yet incorrect configurations that satisfy sensing constraints but deviate from the true spatial geometry. To mitigate this issue, we propose Clustered Localization via Variance-weighted Edge Reinforcement (CLOVER), which forms robust local clusters, derives coarse inter-cluster geometry from lightweight hop-based cues, and refines the global layout via variance-weighted graph optimization. CLOVER enforces global geometric consistency while maintaining efficiency by compressing nodes into clusters and progressively stitching them at the cluster level. Evaluation results show that CLOVER preserves localization accuracy and geometric stability under noisy and sparse deployments.
