The project Robotic companion exploiting affective feedback for modelling emotional state of the patient and adapting the rehabilitation treatment, co-funded by the Italian Institute of Technology and in collaboration with the Department of Bioengineering of the Politecnico di Milano and the Maugeri Foundation, is aimed at studying the possibility to model the affective state of a patient involved in a robotic-assisted rehabilitation therapy session.
In particular, we would like to recognize states like stress, involvement, boredom, from physiological data taken by non-invasive sensors and from other data sources, such as video images taken during the session. The final goal is the improvement of the rehabilitation behaviour of the robot, which will be adapted to the affective state of the patient. We have defined protocols for data acquisition and actually performed data acquisition on the partner site, Fondazione Maugeri. We are also implementing new robots to improve the experimental activity.