A soft pneumatic robotic finger integrated with sound sensor for texture identification of cloths through friction.

Abstract:
The automation of the fabric industry is much behind compared to other industries. Among other reasons, the inability of the widely used, rigid robotic grippers to handle sensitive materials like textiles, is one of the most important. In this paper, a 3 Degrees of Freedom (DOF), pneumatic finger is developed, aiming on touching and identifying textiles with different textures. The concept originates directly from the human finger’s anatomy, its movements and unique sensing abilities. Its design and dimensions are based on an average human finger sampled from 20 different specimens, both men and women randomly chosen. The finger is composed by three independent hollow chambers and actuated by air. It is very flexible and fabricated by silicone rubber. The joints’ movement is enabled by inserting compressed air into each of the independent hollow chamber. The ability to “sense” is developed by a sound sensor at the edge of the fingertip. The fabrication process, kinematic analysis, and a prototype of the proposed finger are presented in this paper, along with the resulting signals received by the sensor, while being rubbed against 10 different textiles. Despite the low roughness of the selected textiles, the prototype demonstrates efficiency in the identification process similar to that of human’s.