A constrained localization algorithm for improving the efficiency and accuracy of forming curved hull plates in shipbuilding. Issue 6 (3rd June 2019)

• Record Type: Journal Article
• Authors: Zhao, Bao; Wu, Zhouqi; Zhen, Xijin; Xi, Juntong
• Is Part Of: International journal of computer integrated manufacturing. Volume 32:Issue 6(2019)
• Page Start: 615
• Page End: 627
• Publication Date: 2019-06-03
• Subjects: Constrained localization algorithm -- curved plate forming -- searching correspondences -- iterative closest point; Computer integrated manufacturing systems -- Periodicals; 670.427
• DOI: 10.1080/0951192X.2019.1599429
• Languages: English
• ISSNs: 0951-192X

Abstract

ABSTRACT: Localization with edges constrained is an effective method for improving the efficiency of fabricating curved plates. The iterative closest point (ICP) algorithm is one of the most widely used methods for localization, while the original ICP method cannot directly be applied to the localization under the edges constrained. In this paper, a two-stage localization approach (the edges are aligned first, and then the whole curved plate is aligned under the edges constrained) is proposed. In the first stage, only the constrained edges are aligned for achieving the optimised localization of them, which can accelerate the convergence of the whole algorithm and improve the accuracy of the aligned edges. In the second stage, the whole curved plate is aligned under the edges constrained. In this stage, a point-to-projection technique is used for efficiently searching correspondences in the ICP frame. Furthermore, a novel double-transformations technique for computing the transformation matrix is also proposed in this stage for ensuring the accuracy of the constrained edges when aligning the whole plate. The results of extensive experiments show that the proposed algorithm has high accuracy and efficiency, and outperforms the existing methods by a large margin.