Abstract: As robots enter everyday life and start to interact
with ordinary people, designing a robot, that is able to perform
perception, becomes increasingly important. Furthermore, to
help a human being in such situations, robots need to receive
human orders and optimize their behavior. Autonomous robot
should make a proper mutual communication between
perception and emotion for creating a desired interaction and
sensation with humans.
This paper presents an approach to the design of an architecture
for human-robot interaction based on the growing understanding
of the neuronal mechanisms and psychological processes
underlying social perception, cognition and action. Interaction
modalities include analysis of speech and of non-verbal human
activities like gestures and general body postures. The proposed
architecture takes into account perception uncertainties,
motivation in human behavior and typical human emotions.
Eventually, models will be evaluated on a physical, autonomous
robot in a typical domestic environment, with some uncertainty.