Research On Probe Radius Compensation Technology For Complex Curved Surface Measurement

Considering China’s rapidly developing manufacturing industry,higher processing and measurement requirements have been proposed for complex surfaces with rapid changes in local curvature.The optical pen visual measurement system is widely used for the measurement of such surfaces,owing to its extensive measurement range,simple operation,and portability.The coordinates of the contact point between the probe and surface are based on the compensation of coordinates at the probe center,which immediately affects the accuracy of the overall measurement system.This paper investigates an effective radius compensation method for utilization in an optical pen vision measurement system.Firstly,the camera imaging model,measurement principles of monocular and binocular cameras,and optimization of the vision measurement algorithm are examined.Furthermore,this study investigates the compensation principles of several existing probe radius compensation algorithms,conducts compensation experiments on various characteristic surfaces,and lists the measurement error sources for several compensating methods.The results provide a theoretical framework for upcoming error reduction.Secondly,the problem of poor compensation and low accuracy in existing probe radius compensation methods is addressed via random sampling during experiments.Subsequently,a mesh-spline compensation algorithm is proposed,which involves using a quadrilateral grid to mesh the scattered points from random measurements to obtain the topological relationship between the center and the obtained surrounding points.The topo logic paper describes the key parameters of the grid and the generation method in detailing and building a rectangular topological grid.The identified topological mesh is used to invert the B-sample surface control vertices and perform NURBS surface interpolation fitting to obtain the normal vectors of the mesh centroids,consequently compensating for the probe’s contact sites with the surface.The impact of rapid curvature changes on the measurement system can be reduced by fitting complex free-form surfaces with spline surfaces.In addition,this paper introduces various surface point cloud fitting algorithms,measures several complex surfaces typically found in the industry,and compares the compensated and theoretical point clouds to evaluate the proposed algorithm’s precision and accuracy.Finally,the experiments demonstrate that the proposed algorithm compensates for the point clouds obtained under random sampling at the micron level,which addresses the accuracy requirements in the industrial field and facilitates its application in field measurements.