Artificial skin using photothermal optical fibers for material identification

This paper presents the development of an artificial skin that can identify material of objects via temperature changes upon contact. By using a cobalt-doped photothermal optical fiber as the heat generation element and fiber Bragg gratings (FBGs) as temperature sensing elements, the artificial skin monitors the heat transfer during contact, providing insights into the temperature profiles and material properties of an object. The efficacy of the artificial skin is tested and verified on three distinct materials: metal, glass and wood and an artificial neural network (ANN) is developed for material classification. The ANN demonstrates 100% recognition accuracy by analyzing the thermal evolution of the FBGs within 5 seconds of contact. This novel material recognition approach offers an alternative domain for tactile recognition. When combined with the acquisition of physical attributes, it allows for more reliable and intuitive object recognition, paving the way for advancements in robotic perception and interaction.