Research On Several Issues Of Soccer Robots

Robot Soccer game is a high-tech oppositional activity that develops rapidly in recent years, it is the combination of the popular physical culture and hi-tech. Robot soccer including many fields such as robotics, intelligent control, wireless communication, image processing, mechanics and so on. We establish our school's robot soccer team, designs and implements a decision-making system for theapplication in the middle-size league of RoboCup-The World Cup of SoccerRobots, and does research on the robot's path planning methods.The main work of the thesis is as follows:The robotics has a good carrier and algorithm can be easy tested good or bad when we have established soccer robot hardware platform. The soccer robot hardware platform including two important structures which are omni-vision system and omni-directional motion structure. We can obtain abundant and precise vision information, because the omni-vision system can see 360 degrees direction of the environment information. Omni-directional motion structure including three omni-wheels which dnved by electromotors, and it can freely travels in a confused dynamic circumstance and achieve an accurate positioning in a competition.We have designed and implemented robot soccer decision-making system by using states space and ratiocinative regulations. The decision-making system maintains a finite state machine, and behaviour selection method which is based FSM (finite state machine) realizes behaviour planning. Through decomposition of specified tasks, the soccer robot can coordinate multi-behaviour under dynamic environment and react rapidly. This decision-making system is used in our soccer robot and attends the 2005 China Robot Match.Dynamic information model is used to describe the motion characters of soccer robots. Through analyzing the motion states of soccer robot, we establish a new dynamic information model The new dynamic information model has a better description of influences which are engendered by soccer robot's movement. We obtain a new path planning method by combining the new dynamic information model and linear programming navigation gradient method. There are many merits of the new path planning method, including quickly implement speed, avoid local minima. It takes a good effect in the experiments.