Real-time Image Localization and Dynamic Path Planning for Middle-Size Soccer Robots

• Main Authors: Wei-Sheng Chen, 陳偉生
• Other Authors: Tzuu-Hseng S. Li
• Format: Others
• Language: en_US
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/99201681753307522885

碩士 === 國立成功大學 === 電機工程學系碩博士班 === 94 === This thesis is mainly to confer the study of real-time self-localization, dynamic attacking/defending path planning, and dynamic tracking for the middle-size soccer robots. According to the middle-size robot soccer league of RoboCup, the middle-size soccer robot (MSR) must be a fully autonomous robot. For our MSRs, the omnidirectional vision system with broad view is mounted on the top of the robot. By using the fast image processing system, our robots can analyze the information about the surroundings and identify the positions of important features in the field. Through the real-time image information and the self-localization algorithm, the absolute coordinate and the heading direction of each robot in the soccer field can be calculated, and the absolute coordinate of the ball in the soccer field can be also obtained based on the relationship among robots and the ball. Combining the information of the ball and the locations of the attacking and defending goals, the best attacking or defending path can be determined by different behavior modes. And then, the MSR accomplishes its own behavior by the FLC system. Finally, the efficiency and feasibility of the proposed system are demonstrated by the practical experiments.