HR-HEC: Heuristic-to-Refined Hand-Eye Calibration via Congruent Matching Sets
Li Liu, Hao Wu, Lin Hua, Dahu Zhu
Abstract
Accurate and efficient hand-eye calibration (HEC) is crucial for vision-guided robotic machining system, particularly in precision grinding blades. However, existing closed-form solutions are highly susceptible to noise and outliers, while iterative solutions depend on good initial estimates. To overcome these limitations, we construct a heuristic-to-refined HEC (HR HEC) algorithm that eliminates dependence on initial estimates and calibration rigs. Leveraging an improved 4-Points Congruent Sets (4PCS) algorithm for heuristic search, HR-HEC robustly establishes accurate point correspondences between initial point clouds, and then constructs congruent matching sets (CMS). These sets, together with global point clouds, are transformed into a unified robot coordinate frame for refined search. The globally optimal hand-eye matrix and stitching results are obtained by minimizing the point-to-plane distance. Calibration experiments demonstrate that the blade absolute stitching error (disN) by HR-HEC can reach 0.06 mm, representing 65.9% improvement over state-of-the-art methods, and meeting the accuracy requirements for blade grinding.
BibTeX
@inproceedings{ral2026_hrhecheuristicto,
title = {HR-HEC: Heuristic-to-Refined Hand-Eye Calibration via Congruent Matching Sets},
author = {Li Liu and Hao Wu and Lin Hua and Dahu Zhu},
booktitle = {RA-L 2026},
year = {2026}
}