Research And Design Of The Vision System Of The Full-autonomous Soccer Robot

As a research subject, robot covers a lot of scientific and technology fields. Especially the automobile robot, which is one of its branches, needs higher technology and has extensive application foreground. The vision subsystem, as an important approach of acquiring environment information, is critical to the performance of the whole system. In the RoboCup competition, although the object is known,because of its rivalry, the vision subsystem must be robust enough real-time to insure the rapid reaction to changes of the environment and accurate enough to measure the distance robots and objects and robust enough to adapt itself to different environment, all of these are great challenges for the design of vision subsystems. Taking Ocean University Of China omni-directional mobile RoboCup robots as research objects,this paper has an intensive research into its architecture, vision system ,decision-making system, and control system. The main research work is listed as follow.First, to start with the application background and the development, some related concepts about this subject are introduced, as well as the origin, classification and the temporary situation. Computer vision, robotics vision technology and the development trend of robot vision are also discussed here.Secondly, some introduction about the vision system of the RoboCup middle-sized robots is stated. We design an omni-directional vision in accordance with the requirements of middle-size soccer robot, then environment information be acquired with a field of view up to 360 0 through this vision system .We research the hardware of the vision system, such as omni-directional vision transducer and image acquisition card.Thirdly, a simple description about robot vision is stated, the detailed discussions about the vision processing of robot vision system are given, including image capture and transform, color image preprocessing, color image segmentation, image recognition and object tracing and so on. Based on lots of experiments, comparing to the color spaces, we design the system on the basis of HSV color space. The color segmentation algorithm based on color space threshold, described the target feature extraction and classification description of the basic method, which focused on the goal of the same shape and moments of heart bilinear interpolation method of calculation. By using the feature matching method, the target color histogram analysis and the probability distribution of goals color features; we can get the color of targeted regional distribution center of mass. We can fast and accurately get the tracking results, completion of the positioning and tracking of targets.The next, the introduction of the vision system software of middle-sized robots, which designed by author, is given. We design our vision system based on Omni-vision and Front vision. Based on the detailed discussions of robot vision system, program designing of vision system, such as image capture, color segmentation module, target recognition and target tracking module, are introduced.Finally, the paper concluded, and discussed the need for further in-depth study of the work.