Development and evaluation of emotional robots for children with autism spectrum disorders

Abstract

Individuals with Autism Spectrum Disorders (ASD) often have difficulty recognizing emotional cues in ordinary interaction. To address this, we are developing a social robot that teaches children with ASD to recognize emotion in the simpler and more controlled context of interaction with a robot. An emotion recognition program using the Viola-Jones algorithm for facial detection is in development. To better understand emotion expression by social robots, a study was conducted with 11 college students matching animated facial expressions and emotionally neutral sentences spoken in affective voices to various emotions. Overall, facial expressions had greater recognition accuracy and higher perceived intensity than voices. Future work will test the recognition of combined face and voices.