Robot-assisted rehabilitation systems have shown to provide controlled, quantifiable, and repeatable movement to the patients. However, detection of emotion of subjects during robot-assisted rehabilitation has often been neglected. It has previously been shown detection of emotional state of the patients are the key factors for successful rehabilitation.
Emotion recognition is a rapidly growing area in the human – machine (robot) interaction research. Researchers have been using subject’s face, voice and/or gestures for emotion recognition. In recent years, inclusion of the emotional state of the patient in the control loop has been started in the robot assisted rehabilitation systems. Note that the physiological signals (responses) can also be used for emotion recognition. Emotions trigger the autonomous nervous system responses such as heart rate and breathing period. It is possible to detect the emotional states of the subjects during the rehabilitation, change the difficulty level of the therapy, provide optimum therapy for each subject, and as a result provide more efficient rehabilitation process.
The aim of this project is to develop a system that can change the difficulty of the rehabilitation task adaptively by looking at the emotional state of a participant, and improve the satisfaction of the participant by providing proper rehabilitation exercise with a robot-assisted rehabilitation system called RehabRoby.