Interpretation of the Modality of Touch on an Artificial Arm Covered with an EIT-based Sensitive Skin
During social interaction humans extract important information from tactile stimuli that can improve their understanding of the interaction. The development of a similar capability in a robot will contribute to the future success of intuitive human–robot interaction. This paper presents a thin, flexible and stretchable artificial skin for robotics based on the principle of electrical impedance tomography. This skin, which can be used to extract information such as location, duration and intensity of touch, was used to cover the forearm and upper arm of a full-size mannequin. A classifier based on the ‘LogitBoost’ algorithm was used to classify the modality of eight different types of touch applied by humans to the mannequin arm. Experiments showed that the modality of touch was correctly classified in approximately 71% of the trials. This was shown to be comparable to the accuracy of humans when identifying touch. The classification accuracies obtained represent significant improvements over previous classification algorithms applied to artificial sensitive skins. It is shown that features based on touch duration and intensity are sufficient to provide a good classification of touch modality. Gender and cultural background were examined and found to have no statistically significant effect on the classification results.