基于嵌入式的机器人视觉伺服寻迹系统研究-A SELF-LOCALITION METHOD WITH MONOCULAR VISION.pdf

A SELF-LOCALITION METHOD WITH MONOCULARVISION FOR AUTONOMOUS SOCCER ROBOT LIU BING-RU, XIE YUN, YANG YI-MIN, XIA YI-MIN, QIU ZHEN-ZHEN Faculty ofAutomation, Guangdong University ofTechnology, Guangzhou 51 0090,China E-MAIL: lbingrug2 1 , , echosun80@ Abstract: processing system. However, a overall vision is prohibited In autonomous robot soccer system, every autonomous soccer robot is an unattached intellective robot. The robot has using in robot soccer matc. In order to make system know its decision subsystem, communication subsystem, vision and the information of overall situation on site, one of the key multi-sensor subsystem, machine-electric control subsystem problems must be solved is the self-location of soccer etc. But any overall vision is forbidden. To get the overall robot. information of the field, the self- localization of autonomous Location is to determine the position in 2 dimension soccer robot is one of sticking points and must be solved. In this paper, a geometrical self-localization method of working environments relative to the position of the autonomous soccer robot is introduced. There is a coordinate of overall situation 21. The main self-location precondition that is some color landmarks in the field have method with monocular vision for autonomous soccer robot been recognized by cyber-vision. A method of depth is inertia location and land marked location. Land marked measurement by monocular vision is used in the geometrical locaion bases on robot vision, uses some known land mark self-localization method. And three possible factor bring in the robots located enviro