Segmental Sensor Weighting Accuracy Evaluation Method for Fiber Optic Shape Sensing

An accuracy evaluation method for optical fiber shape sensing is proposed, which allows us to compare the performance of sensors with different specifications. To achieve a more accurate assessment of the curve reconstruction, we fully consider the spatial information of the reconstructed curve and perform flexible segmental weighting on the shape sensor. Subsequently, by normalizing the average sensing length, the performance of sensors with different specifications can be compared. During numerical simulation, we reconstruct spatial curves using four sensors with different specifications, then, evaluate the accuracy of the reconstruction results by using the traditional accuracy evaluation method and the proposed method, respectively. The results show that the sensor reconstruction accuracy from the global and local scales of a single sensor can be evaluated, meanwhile, the normalized errors of sensors with different specifications are not always optimized with the increase in the number of sensing points, which can be used as a reference to make a balance between the increase in sensor cost and the improvement in reconstruction accuracy. In addition, we demonstrate through experiments that the proposed shape reconstruction accuracy evaluation method has higher robustness than traditional methods.