Decision-making and simulation in multi-agent system based on neural network and PSO

This paper proposes a method using neural networks and particle swarm optimization (PSO) for the decision-making in the multi-agent system. In this paper, a neural network is used for behavior decision controller The input of the neural network is decided by the last strategies of other agents. The output determines the next strategy that the agent will choose. The connection weight values of this neural network are encoded as fitness, and the weight values are updated using the particle swarm optimization algorithm. Here, the weight values imply the adaptiveness of agents in multi-agent system. The validity of the decision model is verified through simulation. Experimental results show that the decision model can successfully simulate agents' dynamic learning and make agents choose the appropriate strategies.