Through robotic soccer issue, we focus on “perception” and “situation and behavior” problem among RoboCup physical agent challenges [1]. So far, we have implemented some behaviors for playing soccer by combining four primitve processes (motor control, camera control, vision, and behavior generation processes)[2]. Such behaviors were not sophisticated very much because they were fully implemented by the human programmer. In order to improve the performance of such behaviors, we have applied a kind of learning algorithm during off/on-line skill development phase. For example, to acquire purposive behavior for a goalie, we have developed a robot learning method based on system identification approach. We also have developed the vision system with on-line visual learning function [3]. This vision system can adapt to the change of lighting condition in realtime.