A Behavior-Learned Cross-Reactive Sensor Matrix for Intelligent Skin Perception

Abstract

Mimicking human skin sensation such as spontaneous multimodal perception and identification/discrimination of intermixed stimuli is severely hindered by the difficulty of efficient integration of complex cutaneous receptor-emulating circuitry and the lack of an appropriate protocol to discern the intermixed signals. Here, a highly stretchable cross-reactive sensor matrix is demonstrated, which can detect, classify, and discriminate various intermixed tactile and thermal stimuli using a machine-learning approach. Particularly, the multimodal perception ability is achieved by utilizing a learning algorithm based on the bag-of-words (BoW) model, where, by learning and recognizing the stimulus-dependent 2D output image patterns, the discrimination of each stimulus in various multimodal stimuli environments is possible. In addition, the single sensor device integrated in the cross-reactive sensor matrix exhibits multimodal detection of strain, flexion, pressure, and temperature. It is hoped that his proof-of-concept device with machine-learning-based approach will provide a versatile route to simplify the electronic skin systems with reduced architecture complexity and adaptability to various environments beyond the limitation of conventional "lock and key" approaches.