This paper mainly concern about the development of an SOPC based image and control strategy system for a small size humanoid robot. The main function of the image system is pattern recognition. Several geometric figures and the direction of arrows can be recognized in a complex environment at the same time. The image information is captured by the CMOS sensor. The colors and noise will be filtered out first. Moreover, the edge detection approach is utilized to find the feature points so as to analyze the image. The control strategy conforms to the laws of Penalty Kick and other events of the FIRA competition. We describe the methods of the offense and defense modes and other strategies. Our robot can track specific objects via the CMOS sensor controlled by two small servomotors. The motions of the robot are decided by analyzing the condition of the environment from the captured image. Finally, the experiment results show the capability of pattern recognition and the efficiency and validity in the HuroCup of the FIRA.
|Number of pages||6|
|Journal||Journal of Harbin Institute of Technology (New Series)|
|Issue number||SUPPL. 2|
|Publication status||Published - 2008 Jul 1|
All Science Journal Classification (ASJC) codes