|
In recent years , the intelligent robot soccer has become one of the hot research topic in the field of artificial intelligence and intelligent control . Robot soccer system is a real-time dynamic , confrontational highly complex environment , it relates to robotics , mechatronics technology , communications and computer technology , robot vision and sensor fusion technology , intelligent control and other high-tech , has attracted many domestic and foreign The experts and scholars to study it. Robot soccer game , both results of these studies show platform confrontation , but also a high-tech platform . Behavior and path planning is an important part of the intelligent robot soccer system research , in order to obtain good results in the game , win , we must have good behavior planning , but also have a good path planning , only the combination of both can be more accurate , more rapid completion of the offensive , shooting sports purposes . This paper focuses on robot the shot behavior and path planning , research method of shooting for the shot data underutilized , only for the current situation has been improved , and explore efficient algorithms to solve problems in path planning . Improvements for shooting behavior is mainly focused on the full use of the saved shot data , processing the data through an adaptive neuro-fuzzy system based on the class of Gaussian function and fitting out the best shot players , goalkeeper and shot point between a function , and the simulation study , the results show a very good way to improve shooting efficiency . Path planning problem , first introduced the three commonly used path planning method , and compare the advantages and disadvantages . Then proposed path planning method based on the S- adaptive genetic algorithm , the main innovation is that the change in the genetic operations of crossover and mutation probability , be able to make it with the autonomous individual changes in the genetic algebra and groups change the probability value can better preservation of the effective individual , makes the algorithm faster convergence to obtain optimal individual under the same conditions , the path of the solution in the S - adaptive algorithm achieve obstacle avoidance behavior in a short time , and the path is superior to the conventional genetic algorithm .
|