High-precision metrology and digital twin reconstruction for large-scale curved surfaces: a metrological laser radar based calibration approach

Abstract Metrological laser radar (MLR) systems have demonstrated significant potential for improving the measurement efficiency of large-scale components, such as aircraft wings, fuselages, and antenna surfaces. However, the lack of traceable evaluation methods for point-cloud-based contour parametrization remains a critical challenge. This study addresses this gap using three key innovations. (1) Traceable coordinate measurement framework: a self-developed large-scale curved surface reference device was built by using an optimized traceable coordinate measurement method to establish a metrological chain for error compensation and uncertainty propagation of the assembled concave and convex surface. (2) High-precision digital twin modeling: a partitioned non-uniform rational B-spline (NURBS)-based reconstruction algorithm was proposed, reconstructing a surface digital model with the deviation root mean square (RMS) of the contour point being within 0.01 mm. The fidelity of the model was rigorously validated through uncertainty analysis and experimental repeatability verification. (3) Evaluation of laser radar measurment capability: a comprehensive evaluation experiment was carried out to assess the contour measurement capabilities of the MLR across distances from 1 m to 10 m. Results demonstrated that the laser radar achieved a contour error RMS of 0.234 mm at 2 m, 0.315 mm at 5 m, and 0.250 mm at 10 m for concave profiles, and 0.259 mm at 2 m, 0.226 mm at 5 m, and 0.242 mm at 10 m for convex profiles. This work provides an evaluation method of the contour measurement capability for MLR systems, guaranteeing industrial quality in aerospace and automotive manufacturing.