A Novel Shape Reconstruction Technique Using Flexible Differential Inductive Sensor Array
Abstract
Accurate shape sensing is crucial in robotics, biomedical engineering, and structural health monitoring for tasks like robotic guidance and infrastructure assessment. However, challenges such as dealing with complex geometries during sensing and data acquisition while maintaining high accuracy remain. This study introduces a flexible differential inductive sensor array for 2-D shape sensing and reconstruction. The system employs two flexible printed circuit board (FPCB)-based inductive sensor strips, separated by a fixed distance, to measure curvature through relative inductance shifts. The differential design suppresses background noise and improves the signal-to-noise ratio. A 3-D finite element model (FEM) was developed for feasibility and parametric studies that led to sensor array optimization. Simulation results showed that a smaller spacing distance of transmitter and receiver coils extends the dynamic range of measurable curvature, while larger spacing enhances resolution for larger curvature measurements. For experimental validation and demonstration, a prototype system was fabricated, incorporating a 3-D-printed flexible mount, a miniaturized circuit system, and shape reconstruction algorithms. The results show that the system can measure the arc shape with an average maximum positional error of 1.97 % for curvature radii ranging from 40 to 100 mm, and 2.45 % for radii ranging from 40 to 400 mm. The system was also tested on two irregular shapes, achieving an average error of 1.25 %. Its capability was further demonstrated through 3-D geometry reconstruction of a dented pipe structure. These results highlight the system’s high accuracy, superior resolution, and great adaptability for various potential applications, offering valuable insights and advances for optimizing shape sensing.