Flagg, A., Tam, D., MacLean, K. E., and Flagg, R., “Conductive Fur Sensing for a Gesture-Aware Furry Robot,” in Proceedings of IEEE Haptic Symposium (HAPTICS '12), Vancouver, Canada, March 2012, pp. 99-104.
Recent advances in artificial intelligence suggest that machines will soon be capable of communicating in ways previously considered out of their reach. For example, humans engage in sophisticated emotional communication through the language of touch. What technical capabilities would enable computers to do the same? As our group examines this question in the context of emotional touch between a person and a furry social robot, we require sensors designed to detect and recognize subtle, nuanced touches. To this end, we demonstrate a new type of sensor based on conductive fur, which is sensitive to movements unavailable to conventional pressure sensors. The sensor captures motion by measuring changing current as the fur's conductive threads connect and disconnect during touch interaction. We then use machine learning to classify gestures from this time series. An informal evaluation with seven participants found 82% recognition of a 3-gesture set, showing promise for this approach to gesture recognition, and opening a path to emotionally intelligent touch sensing.