FLASH™ arouses positive emotions and people feel safe with him. They recognize that his appearance as well as his behaviors, are human-like.
Even though his head seems to play the most important role during interaction, body posture is also assessed, especially in the case of dynamic behaviors. It is natural to assume that interaction with a social robot should follow human-like guidelines.
The design of a social robot faces a number of challenges, like consistency of appearance and behavior, perception, interaction and emotional expression.
The employment of a balancing platform resulted in obtaining natural, fine and smooth robot movements, well perceived by humans. Furthermore, in contrast to many multi-wheeled platforms, a two-wheel balancing platform with high positioned center of gravity can easily move on rough ground and overcome slopes.
Structure of an arm resembles that of its human counterpart and has seven degrees of freedom. Each hand consists of a total of four fingers with three degrees of freedom each.
His vision system is comprised of a hi-res camera mounted in his head and a Microsoft Kinect device playing the role of an advanced motion sensor. It allows him to detect humans as well as provide information on position of particular elements of their bodies. It also provides gesture recognition algorithms, 3D face tracking, and recognition of some facial features.
Both video streams can be processed by algorithms which provide: basic image processing (e.g. blurring, thresholding and morphological operations), object detection (e.g. human faces or certain body parts), color and movement detection, and much more.
FLASH™ can connect to the Internet and browse news services, send and receive emails, check weather forecasts, use your Facebook account, Google calendar, etc. Information gathered by the robot could potentially be used to stimulate everyday human-robot interaction.
Perhaps the most advanced available affective mind, FAtiMA (FearNot! Affective Mind Architecture), based on the Orthony, Clore and Collins appraisal theory of emotions, was also successfully integrated with EMYS' control system. The decision system can also be assisted by a human operator.
Thanks to the integrated computational model of emotion with appraisal capabilities, information gathered from the Internet like news, weather forecasts, messages, etc. can be affectively assessed to extract the emotional meaning.