Analyses of information flow in the cycle of action and perception in an object-pursuit task of the humanoid robot

Biologically-inspired robots have been studied with respect to biological intelligence which can provide complex behaviors. The humanoid robot is a trend that has been used to investigate how the human body works (bipedal walking, for instance). In this research, however, we focus on on-demand planning and rescheduling within a dynamic environment. In the object pursuit task, experimental results demonstrated a online circulation between the internal controller and the environment to prevent fluctuations caused by an irregular noise, and to enable the robot to reach the target position without the endless loop of the identical motions that frequently occur in real time robotics.