Influence of Reverse Geometric Modeling parameters on Sensor model Accuracy in Diagnostic Systems

Accurate alignment of sensors in robotic diagnostic systems is essential for reliable data acquisition and overall system performance. However, the accuracy of sensor positioning often depends on the precision of the reverse-engineered CAD models used for hardware integration. Despite the wide use of reverse geometric modeling in practice, the influence of its parameters on the final model accuracy remains insufficiently explored. In reverse modeling, the precision of the reconstructed geometry is strongly influenced by software settings such as the type of surface fitting algorithm and the degree of triangular mesh reduction. Inaccurate parameter selection can lead to geometric deviations, which subsequently affect sensor calibration and mechanical compatibility within diagnostic systems. This study investigates how different reverse modeling methods and parameters affect the dimensional accuracy and integrity of CAD models. The analyzed factors include surface fitting methods, dynamic sectioning for reconstructing geometries, and triangular mesh reduction. A physical sensor mount was digitized using high-resolution optical 3D scanning and processed with various CAD and CAQ modeling strategies. This sample was basic cuboid sensor with dimensions 100 x 10 x 10 mm with 8 mm diameter. A simple cuboid sample was chosen to study planar and cylindrical surface fitting: freeform surfaces will be tested in next experiments. Experimental results identify the most effective reverse modeling approaches for sensor-related applications and highlight how mesh decimation impacts accuracy, particularly in areas with high curvature or fine details. The findings provide practical guidance for designers and engineers working on sensor calibration and integration in automated diagnostic systems.