For my Master's thesis project, I collaborated with Dr. Karon MacLean at the University of British Columbia to explore the design of low-cost, low-tech touch sensing hardware, and its applications in haptic artificial intelligence. Done as part of the Haptic Creature project, our focus was designing a furry animal-like machine to intelligently recognize emotional touch gestures using machine learning methods. We were inspired by Perner-Wilson's wonderful thread stroke sensor (in our smart fur design), and Schmeder and Freed's fascinating piezoresistive fabric pressure sensor. The ultimate goal of this work is the design of more emotional, potentially therapeutic machines that can help people feel better.
See Cuddlebot on MIT Tech Review or on the cover of the New York Times Magazine, Innovations issue.
Exploring low-tech touch sensing
Construction Summary: Body, pressure sensing, fur sensing
Cuddlebot at the TEI Conference: Barcelona, 2013
Piezo-resistive fabric sensing: Adaptation of Schmeder & Freed
Gesture data: Pressure, position, fur signals