Skip to Main Content
This paper proposes a perceptual system based on computational intelligence. Basically, a robot should perceive its environment based on the physical body. This indicates the meaning of a sensory input can be interpreted according to the features of the physical body. First of all, a robot performs map building according to the physical size of the robot, and then, the robot extracts structure of color information form the environment, that can navigate the robot to a target point. Finally, the robot learns collision avoiding and target tracing behavior based on visual perception.