Implementation of Dynamic Faces Based on Proxemics for Robot-ASD Children Interaction

This paper shows how a robot can be used to simulate facial expressions and, thus, increase the interaction between it and autistic children. These facial expressions are based on proxemics which uses information taken from a laser sensor. Following a proposed algorithm, the robot can choose an adequate face animation, expressing emotions in several degrees. The capability of displaying emotion that changes throughout the time with different intensities enhances the interaction between autistic children and socially affective robots, which can be applied as a support tool in their therapy. Moreover, once the robot acquires the proxemic information, it can change its behavior in an appropriate way to try catching the child's attention and interest. In this context, as both movement and facial expressions play an important role in this interaction, this way of processing emotion and displaying dynamic faces can bring a more natural and improved interaction.