Shape sensing using neural network of radial basis function based on Fiber Bragg Grating

Abstract Shape sensing using Fiber Bragg Grating can be applied in many fields such as medical and robotics, and the accuracy of its reconstruction is the key to current research. At present, the application research in two‐dimensional planes has reached a high accuracy, how to improve the accuracy of three‐dimensional space reconstruction is the current challenge. An interpolation method using Radial Basis Function neural network is proposed for curve reconstruction in three‐dimensional space, which is compared and analyzed with the existing commonly used cubic‐spline interpolation method. Through theoretical validation and experimental verification, both the tip error and the average distance error of the sensor are reduced.