Research on the Decision Model of Multi-agent cooperative confrontation
Zhongqiu Zhang · 2020
Multi-agent systems are an important area of distributed artificial intelligence. Its goal is to turn large and complex systems into small systems that communicate and coordinate with each other and are easy to manage. Cooperative confrontation has been widely used in different fields such as production scheduling and virtual simulation. Therefore, it is of great significance to study the decision model of multi-agent cooperative confrontation. In this paper, with the aid of a multi-agent cooperative confrontation system, the method of Q-Learning algorithm and value function approximation is used to realize the path planning of the agent. We simulated an experiment to show the application effect of reinforcement learning algorithms in path planning, and compared the convergence of MC Control, SARS, and Q-Learning algorithms, which illustrates the advantages of Q-Learning. Then, simulation experiments are performed to show the application effect of the reinforcement learning algorithm combined with the value function approximation in path planning.