A Tactile Glove Based on Capacitive Pressure Sensor Array for Object Shape Recognition

The tactile glove based on a pressure sensor array plays an important role in the field of tactile perception. However, fabricating a large-area, high-density flexible pressure sensor array based on high-performance sensing units on the same substrate remains a challenge. Here, based on cost-effective flexible printed circuit board (FPCB) technology, we have completed the integrated fabrication of 137 electrodes on the same substrate of a capacitive pressure sensor array. The electrode adopts the single-layer interdigital structure, effectively eliminating crosstalk and packaging difficulties. Furthermore, we also fabricated the polyurethane (PU) and the barium titanate (BaTiO3) porous dielectric layer based on nonsolvent induced phase separation (NIPS). The sensing unit exhibited a high sensitivity of 0.141 kPa−1, a wide detection range of 0~190 kPa, a fast response time of 90 ms, and stability, meeting the practical operational requirements of tactile perception in prosthetic hands. Then, we developed an object recognition system based on the tactile glove, which reflects the pressure changes when grasping different objects. With the assistance of an improved convolutional neural network (CNN), the recognition of nine differently shaped objects was achieved with an accuracy rate of 95.14%. Our proposed design method of tactile glove based on capacitive pressure sensing array proposed by us can provide new ideas for the free fabrication of large-area and high-performance sensor arrays, and has potential application value in fields such as intelligent prosthetic hands, intelligent robots, and human-computer interaction.