Research Of Localization System Of Autonomous Soccer Robot Based On Omni-directional Vision

Among research fields about mobile robot, robot localization, which is the basis of autonomous motion and other tasks, is a pivotal technology. It is also a very valuable research topic, which relates to many different fields.In this dissertation, using Voyager-II autonomous soccer robot as a R&D platform, we study the localization problem of autonomous soccer robot systemically, and design a system of localization based on omni-directional vision and geometry- triangle method, including self-localization and object-localization.Firstly, the concept of soccer robot including its research significance and the RoboCup middle-sized robot game are introduced. The introduction about the concept of vision localization, its important research significance is mainly stated. Vision localization of RoboCup middle-sized robot game and the domestic and exoteric progresses are also discussed here.Secondly, Some introduction about the architecture of Voyager-II which is a omni-directional RoboCup middle-sized robots are stated, and its main components such as vision localization system, decision-making system, motion control system and the goal-kick system are surveyed.Thirdly, a simple description about robot vision localization is stated, then the detailed discussions about the vision localization processing of robot vision localization system are given, including image capture and transform,image preprocessing,image segmentation,target recognition and localization and so on.Then next, the introduction of the vision localization system software of middle-sized robots, which designed by author, is given. Based on the detailed discussions of Voyager-II software and vision localization system, the program designing of vision localization system, such as capture of color space threshold module, image process and target recognition module, localization including self-localization and object-localization base on geometry- triangle method module, are introduced. The color segmentation algorithm base on color space threshold,object recognition based on RLE(Runtime Length Encode),noise reduction based on Erosion-Dilation algorithm and localization base on geometry- triangle method are detailed introduced.Facts have proved that the system of localization can be applied to autonomous soccer robot successfully.Finally, the conclusion of the research and some fields for extensive research in the future are shown.