Electrostatic Enhanced Dual-Mode Electronic Skin for Multifunctional Robotic Hands Capable of Object Shape and Material Recognition.

Owing to the increasing requirement for robotic systems to interact intelligently in unstructured and dynamic environments, multimodal perception has become an essential and challenging task. In this study, we introduce a multifunctional robotic hand that can distinguish the shape and material properties of objects using a dual-mode electronic skin (e-skin) capable of non-contact and contact sensing. The e-skin is composed of a polarized expanded polytetrafluoroethylene electret embedded in Ecoflex, thus enabling non-contact sensing via the electrostatic-field effect and contact sensing based on the triboelectric effect. The embedded electret architecture facilitates a high internal charge density, thereby significantly enhancing the intensity and range of non-contact sensing, an advantage not achievable using conventional approaches. Integrating the dual-mode e-skin into a robotic arm endows it with multifunctional capabilities; furthermore, with the assistance of a long short-term memory neural network, the robotic hand achieves 100% and 97.35% accuracies in object-shape and object-material recognition, respectively. This study demonstrates the potential of the proposed e-skin as a versatile multimodal sensing interface for robotic platforms, thereby advancing autonomous and intelligent robotic interactions.