Multidimensional Parameter Monitoring and Cross-Sensitivity Decoupling in Force-Sensing Optical Fibers: Mechanisms, Algorithms, and Experimental Validation

DOI: https://doi.org/10.62517/jes.202602226

Author(s)

Xin Liu, Zhonglin Xu, Tao He, Hao Xiang, Jinhui Zhao, Yaoyi Jiao, Dai Hou, Junguo Hu, Chen Chen, Qian Wang, Bei Wang

Affiliation(s)

Hubei Siji Technology Co., Ltd., Wuhan, Hubei, China

Abstract

The demand for high-precision, multi-analyte sensing in complex environments has established optical fiber sensors as a cornerstone technology for intelligent robotics, structural health monitoring, and biomedical diagnostics. However, deployment is fundamentally hindered by cross-sensitivity, wherein optical waveguides respond simultaneously to multidimensional strain, spatial multi-axis forces, and temperature fluctuations, rendering the extraction of individual physical parameters an ill-posed mathematical problem. This paper presents an exhaustive methodology for multidimensional parameter monitoring utilizing force-sensing optical fibers. The proposed system integrates an assembled 3-UPU (universal-prismatic-universal) compliant parallel mechanism with Fiber Encapsulation Modules (FEM) to physically decompose six-axis forces into measurable linear strains. To resolve the inherent opto-thermal cross-sensitivity, this study explicitly details two decoupling frameworks: a deterministic sensitivity matrix inversion via mathematical conditioning, and a Physics-Informed Neural Network (PINN) that embeds structural kinematic constraints directly into the optimization loss function. Experimental validations utilizing high-resolution optical frequency-domain reflectometry demonstrate exceptional precision. The sensor architecture restricts maximum Type-I errors to 4.52% of the full scale across six dimensions, while the PINN algorithm achieves a temperature resolution RMSE of 0.4 °C and a strain decoupling error of 15 . This synergistic integration of engineered structural mechanisms and physics-guided machine learning provides a robust, electromagnetically immune paradigm for multimodal tactile perception.

Keywords

Force-Sensing Optical Fiber; Multidimensional Parameter Monitoring; Cross-Sensitivity Decoupling; Fiber Bragg Grating (FBG); 3-UPU Parallel Mechanism; Physics-Informed Neural Networks (PINN)

References

[1]ZHANG L, WANG Y, SUN M, et al. A review of optical fiber sensors for multi-dimensional force sensing in robotic applications. IEEE Sensors Journal, 2022, 22(18): 17450-17465. [2]LIU X, CHEN H, YI J, et al. Research on decoupling algorithm of six-axis force sensor based on fiber Bragg grating. Measurement, 2021, 173: 108665. [3]WANG J, QUAN L, GUO Y, et al. Design and kinematic analysis of a 3-UPU compliant parallel mechanism for multi-axis force sensing. Mechanism and Machine Theory, 2023, 185: 105321. [4]KARN G P, ZHAO L, WIGGINS P, et al. Physics-informed neural networks (PINN) for resolving cross-sensitivity in optical fiber sensors . Nature Communications, 2024, 15(1): 452-468. [5]CHEN Z, ZHANG J, ZOU M. Temperature-strain simultaneous measurement using a single fiber Bragg grating based on transfer matrix method and deep learning. Optics & Laser Technology, 2022, 149: 107842. [6]SMITH A, DOE J. Optical frequency domain reflectometry for high-resolution strain mapping in compliant structures. Journal of Lightwave Technology, 2023, 41(9): 2810-2818. [7]ZHAO Y, LI X, LIU Q. Advanced fiber-optic tactile sensors for Healthcare 5.0: materials, structures, and algorithm. Sensors and Actuators A: Physical, 2025, 368: 115024. [8]HUANG W, LIU G, ZHANG R, et al. Decoupling of multi-dimensional force/torque signals in FBG sensors using a physics-guided neural network. IEEE Transactions on Instrumentation and Measurement, 2022, 71: 1-12. [9]KAPOOR A, SINGH S. Sensitivity matrix conditioning and optimization for multi-parameter fiber optic sensing. Optik, 2021, 242: 167310. [10]PARK S, KIM J, LEE H. Soft silicone-encapsulated fiber optic sensors for large-strain tactile perception in robotic grippers. Advanced Intelligent Systems, 2023, 5(4): 2200315. [11]XU L, ZHOU J, WANG P, et al. High-precision 6-DOF force/torque sensor based on FBG and parallel compliant mechanism . Mechanical Systems and Signal Processing, 2024, 206: 110892. [12]RAISSI M, PERDIKARIS P, KARNIADAKIS G E. Physics-informed neural networks: A deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations. Journal of Computational Physics, 2019 [13]YANG M, ZHANG H, LI Y. Analysis of the thermo-optic and photoelastic effects in FBG sensors under extreme thermal gradients . Photonic Sensors, 2022, 12(3): 255-269. [14]GUO T, LIU F, LIAN W, et al. Recent advances in fiber Bragg grating sensors for structural health monitoring. Sensors, 2021, 21(23): 7915. [15]LIU D, SUN S, GAO X. Curvature and strain sensing mechanism of liquid silicone-housed optical fibers for human-machine interaction . ACS Applied Materials & Interfaces, 2023, 15(12): 15980-15992.