Research On Control Strategy Of Soccer Robot Combined With Winner-take-all

With the emergence and rapid development of AI technology,the country pays more and more attention to the convenience brought by AI.As the integration of multiple intelligent bodies of AI,the research of robots has also been paid more and more attention by researchers.Robots,as a new generation of agents,are full of intelligent “strategies”,making robots more flexible.Strategy design is a very important link in the field of robot control,and a good strategy can help the robot move in a "humanoid" motion.To enable a robot to move completely,it needs all links to cooperate with each other.This requires a flexible function control system,which is the core of controlling all robot motion commands.As an important type of robot,football robots are now being widely studied by scholars.It is a comprehensive set of robots,intelligent control,data fusion,computer technology,wireless communication,image processing,mechanics and other disciplines and technologies.The product itself is a typical nonlinear control object,which provides a very good experimental platform for various control theories.This article mainly introduces the corresponding strategies adopted by soccer robots according to different sports needs in the process of sports.The main research contents are as follows:The background research of robots was launched,and the theoretical basis of robots as products of artificial intelligence advanced agents were introduced,as well as the research significance and development trends.By elaborating the positive impact of football robot research on academia and industry,citing the research object of the article,explaining the research content and significance.Explore the feasibility of WTA theory in the decision-making problem of soccer robot movement,focus on introducing the basic theory of WTA,the advantages and disadvantages of the algorithm and the challenges faced,and comprehensively summarize the WTA theory.By analyzing the application of the theory in multiple real scenes,it is verified that it is widely applicable and technically feasible.Aiming at the needs of football robots in the complex environment of real-time confrontation,which requires a higher ability to cope with the real-time motion process,a set of strategy schemes combined with WTA theory is proposed.Since many existing researches on strategies only focus on the path planning of the single-robot ball control process,they do not consider the problem that the competition between multiple robots leads to the lack of certain key steps and the loss of integrity in the entire movement process of the soccer robot.This paper proposes a new strategy to solve the above problems: The first step is to use a continuous-time nonlinear dynamic competition model to calculate the optimal ball-robbing robot,and to verify the scientificity and practicability of this model through static competition and dynamic competition experiments;effective solution Multi-robot competition problem;the second step uses the improved APF algorithm to achieve motion obstacle avoidance,correct the shortcomings of the traditional APF algorithm,improve the execution efficiency,and make the ball control route with traps more complete and smooth.Through simulation experiments,the final results are analyzed and discussed and the correctness of the theory and the scientificity and practicability of the proposed theory are verified.