High-resolution, High-Speed and Low-Cost Flexible Tactile Sensor Array System

Xudong Hu,ZhongLiang Zhang,Xiaobin Chen,Fei Wang,Lin Shu,Xiangmin Xu,Xiaoming Tao
DOI: https://doi.org/10.1016/j.measurement.2024.115630
IF: 5.6
2025-01-01
Measurement
Abstract:Autonomous robots mainly rely on visual perception to perceive their surroundings, but some tasks cannot be completed or are challenging without a sense of touch. In the design of electronic skins for robots, flexibility, resolution, and rapid response are crucial. In this paper, a high-resolution, low-cost, and high-speed perceptual system is proposed. A sensor array with a spatial density of 455 sensor elements per cm(2) 2 was fabricated. The acquisition system can transmit data at an adequate speed of 5.33 frames per millisecond, in which an improved zero-position method was proposed to eliminate crosstalk. The spatial resolution of the system can reach 0.5 mm, which allows surface feature acquisition. This allows estimation of the grasp stability of the robot. Further, a neural network was designed to recognize the shapes of small objects. In small-object shape recognition, 2400 data from 8 small objects were collected and used for network learning and validation. The accuracy for the validation set was 92.2 %, indicating that the system performed well in distinguishing the sizes and shapes of objects. The system (including the sensor array) costs only $24.41 but exhibits satisfactory performance in key parameters.
What problem does this paper attempt to address?