Intelligent evaluation of fabrics' extensibility from robotized tensile test

Abstract

Purpose The paper aims to propose an approach to intelligent evaluation of the tensile test. A robotized system is used that performs the fabrics tensile test and estimates the extensibility of the samples using a feed‐forward neural network while trying to imitate the human expert estimation. Design/methodology/approach The specifications of the tensile test are derived by an extensive observation of the respective experts' estimation performance. The fabric sample size and the experimental conditions are specified. Linguistic values of the term “fabric extensibility” are extracted through a knowledge acquisition process. The tensile test is performed by a robot manipulator with a simple gripper and the experimental measurements (force, strain) are fed online into a neural network. The network is trained according to the extensibility estimations of the experts. The trained network is tested in estimating unknown fabric's extensibility. Findings The results demonstrate that the system is capable of estimating the extensibility of new fabrics. Originality/value This work can be integrated in the robotized sewing process with intelligent control where the fabric's extensibility in terms of linguistic values is necessary. The proposed system initiates a new approach, in which the fabric properties are expressed and used in a way that will facilitate the introduction of the artificial intelligence methods into the clothing industry.