多代理协作粒子群背包问题求解算法

Collaborative Particle Swarm Optimization Algorithm for Knapsack Problem Based on Multi-Agent

代理技术是分布式人工智能的重要研究领域,其特性非常适应于解决分布式问题,并多个领域成功应用。基于粒子群的社会性特征,本文将代理技术与粒子群进化方法相集合,提出了基于多代理的协作粒子群求解问题模型。通过粒子群之间的协作作用,克服了传统粒子群方法的不足。通过背包问题的算法设计及实验,表明该算法有效地提高了效率和效果。

Agent technology is one of the most important research fields distributing problems due to its special characters and has been applied characteristics of particle swarm, this paper presents collaborative particle of distributed artificial intelligence. It is adapted to solve in many fields successfully. In accordance with the social swarm optimization model based on multi-agent by combining agent technology with particle swarm optimization. Through collaboration between the particle swarms, the model overcomes deficiency of the traditional methods of particle swarm optimization. Through the algorithm applied in the knapsack problem and experiments, the algorithm has been proved to increase the effect and efficiency effectively.