ELTac: A Vision-Based Electroluminescent Tactile Sensing Skin for Force Localization and Magnitude Estimation

Large-area tactile sensing for robotic manipulators is an important capability to enable robots to perceive interactions with environment around them and for intuitive human–robot collaboration. In this article, we introduce a novel vision-based tactile sensing methodology that employs electroluminescent (EL) panels with a deformable soft skin that modulates light intensity based on applied forces to realize force localization and magnitude estimation for multipoint contact scenarios. The tactile sensing module is composed of a transparent rigid skeleton, a sensing skin composed of a thin and flexible EL panel, a deformable translucent elastomer layer with a pyramid pattern and an opaque outer layer. When a force is applied onto the skin, the deformable layer deforms modulating the intensity of light passing through the transparent layer which is detected by a camera embedded inside the module. We utilize image processing, camera models, and statistical fitting to localize single and multiple touch points as well as estimate the magnitude of the forces applied. Finally, the proposed algorithm is tested with five different indenters, and the localization error and the intensity-force mapping are analyzed. A localization accuracy of 6.63 mm has been achieved and normal forces from 3.1 to 9.4 N can be detected with an accuracy of 9.3%–11.7% error range. This work provides a simple and effective solution for the acquisition of position and force magnitude information in human–robot interaction tasks such as guidance and demonstration.