High-Accuracy 3D Shape Sensor Based on Anti-Twist Packaged High Uniform Multicore Fiber FBGs

Abstract

Improving the accuracy of shape sensors based on multicore fibers (MCFs) is challenging but of great importance for real-time 3D shape detection, especially in visually inaccessible areas. In this work, a novel approach is proposed to improve MCF shape sensor accuracy using an ultraviolet transparent liquid mediated fiber Bragg grating (FBG) inscription technique and a twist-isolating packaging method. A newly developed UV index matching liquid (UV-IML) is used to generate uniform light field at all the MCF cores, enabling FBG inscription with high accuracy. Additionally, a new stress fully released (SFR) packaging method is implemented to isolate the sensor from any external twists. The MCF shape sensor shows a maximum relative error of only 3.33% and the lowest reported relative sensitivity error of 1.11% cm⁻¹. Moreover, a real-time 3D shape sensing system with a response frequency larger than 30 Hz is constructed using the unique MCF shape sensor. The highly accurate real-time 3D shape sensing results indicate potential applications for in vivo shape estimation of endoscopies and soft robots.

Graphical abstract

A highly accurate MCF shape sensor for real-time 3D shape detection in visually inaccessible areas is developed. The MCF shape sensor shows a maximum relative error of 3.33% and the lowest reported relative sensitivity error of 1.11% cm⁻¹. The novel ultraviolet index matching liquid FBG inscription method and stress fully released packaging method ensures high accuracy.

Funding

National Naturual Science Foundation of China(T2293751)