A Fingertip Optical Fiber Composite Sensor With Conformal Design for Robotic Perception of Tactile Force

Abstract

This study employs fiber Bragg grating (FBG) sensing for finger force feedback and position measurement, crucial in robotic intelligent perception and human-robot interaction. The fingertip three-axis force sensor placed inside the skinned sensor is threaded to the fingertip cap and finger joint. Such removable conformal configuration offers distinct benefits in compactness and finger composite force measurement. A response surface method (RSM) driven by simulation data is proposed for theoretical modeling and optimization of structural parameters. The experimental results demonstrate sensitivities of 225, 229, and 97 pm/N in <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">x-</i>, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">y-</i>, and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">z</i>-directions within ±10 N, with dynamic measurement errors under 4.84%. Random forest (RF) and back propagation neural network model is developed to classify and regress the finger belly pressure, which achieves 99% classification accuracy and 7.86% maximum regression error in single-point and double-point pressure. Recognition experiments on material texture and international braille have proved its feasibility and reliability. These results are significant to the fingertip tactile force sensor design for human-robot interaction.