论文部分内容阅读
为了克服目前Agent联盟形成策略存在的一些缺陷,提出了一种基于二次招/投标协议的动态Agent联盟形成策略.该策略采用二次招/投标法形成动态联盟,能够在满足子任务时间约束的条件下,充分发挥Agent的并行执行能力;并使得Agent能够动态加入和退出联盟,从而提高了Agent的利用率,保证了自利Agent自身利益的最大化.最后在AgentTool环境下进行协议仿真验证,实验结果表明该协议不存在死锁、活锁和死循环.
In order to overcome some flaws existing in the forming strategy of Agent alliance, this paper proposes a formation strategy of dynamic Agent alliance based on the two bidding / bidding protocols. This strategy forms a dynamic alliance by the method of bidding / bidding twice and can satisfy the sub-task time constraint , And give full play to the ability of Agent to execute in parallel.And make it possible for Agent to join and leave the alliance dynamically so as to improve the utilization rate of Agent and maximize the self interest of Agent.Finally, protocol simulation is carried out in AgentTool Experimental results show that there is no deadlock, livelock and deadlock in this protocol.