Vision System Design And Research On RoboCup Humanoid Robot

With the development of robot technique, soccer robot has become one of the hot focuses in this field. It combines the technique from information, automation, communication and AI and provides a platform to test all these skills.Vision system is the most important apparatus for human being and people acquire most of information from the outside world by their vision systems. In the process of human being's exploration to the nature world, man always hope to find some machine-computing ways to achieve human being's vision function, and to acquire information automatically from outside world. Robot-soccer vision system is just based on chromatic image processing, and with a great demand for its real-time performance, and then it can be an ideal subject for those above research fields.The major content of this paper is to research RoboCup Humanoid vision system. Vision system is a unique way through which the whole system obtains the information of the game. Whether the vision subsystem obtains the information of the game quickly and accurately is the key of the whole game, its capability directly affects the victory or defeat of the game. Vision subsystem needs to process real-time track and recognition for chromatic game image, and realize robot vision functionIn view of research present situation of vision system, combining with our production which has been researched in this system for many years, the major contents of this paper r include:①Vision system color model choice problem studies as well as accurate color identification method. And that the color model chooses is the first steps that the image handles, wait for a shortcoming specifically for the RGB model exposes the deficiency to come coming out in person type robot competition, if drawing accepting light change effect, target color is inaccurate, the main body of a book the characteristic having analyzed the HSI color model, makes use of BP neural networks to carry out color image separation here on the basis.②Humanoid robot optesthesia distance measurement studies. Person type robot optesthesia platform is horizontal can be perpendicular rotate, the robot can gain big range inner environment information under the situation of not moving by rotating of the camera shooting head. Make use of the optesthesia imaging model and image space to change a model, the head turns the lower condition image distance measurement problem by the fact that the right algorithm gets a robot in camera shooting.