A high-accuracy shape sensing method for deformation digital twin monitoring of thin-walled aircraft structures

High-accuracy deformation monitoring of thin-walled aircraft structures, referring to panels of primary load-bearing components such as wings and fuselage sections, is a critical link in surface-shape feedback during flexible manufacturing, as it is a key prerequisite for ensuring compliant assembly. Shape sensing technology provides essential support to achieve this goal. However, traditional methods are often sensitive to strain noise and heavily dependent on material parameters. This paper presents a shape sensing method for deformation digital twin monitoring (DDTM) of thin-walled aircraft structures that combines the inverse finite element method (iFEM) with Bayesian inference and Gaussian process modeling. The method corrects displacement estimation errors online and does not rely on prior material parameters. A bilinear interpolation scheme is used to extend the correction across the full structure under sparse sensing conditions. The approach is tested on reduced-scale wing panels made of aerospace-grade aluminum alloy and CFRP during robot-assisted shape compensation assembly. Experimental results show that the method achieves a maximum relative error of 12.80% and an average relative error of 4.66%. Compared with conventional iFEM, the average error is reduced by 64.42% on average. The proposed method improves sensing accuracy and robustness, providing a reliable and efficient tool for high-precision DDTM during aircraft manufacturing.