We investigate how multipass emotion labelling protocols generate ecologically valid dynamic emotion transition labelling of affective touch behaviour (and other modalities).
The CuddleBot allows us to use an animal model of therapy to examine how touch interactions influence stress and anxiety mitigation.
We develop machine learning models of affect using touch and biometrics (such as EEG signal, skin conductance, heart rate, etc) to support affective interaction with companion robots for health applications.
Our furry robots' breathing is innately calming. Can we use this therapeutically?
Bits are one-degree-of-freedom actuated sketches which we use as a design tool to explore rendering of emotion through physical and visual motion.