论文部分内容阅读
多Agent联盟生成是多Agent系统的关键问题之一,主要研究如何在多Agent系统中动态生成面向任务的最优联盟.为使Agent能稳定的组织起来完成单Agent不能完成的任务并在成本、资源、利益等方面达到一个良好的平衡性能并达到全局最优,提出了联盟多目标综合评价模型,并将量子进化多目标算法应用于多目标多任务Agent联盟问题,运用编码的映射,将资源组合和任务分配合并为一个过程,降低了问题的复杂性.对比实验结果表明该算法求得的解的质量高,平衡性好,能有效避免了联盟死锁和资源浪费.
Multi-Agent alliance generation is one of the key problems in multi-Agent system, and mainly studies how to dynamically generate the task-oriented optimal alliance in multi-agent system.In order to make Agent can be organized to complete the task that Single Agent can not complete and the cost, Resources, benefits and so on to achieve a good balance of performance and achieve the global optimum, a coalition multi-objective comprehensive evaluation model is proposed, and the quantum multi-objective evolutionary algorithm is applied to multi-objective multi-agent Agent alliance problem. By using coding mapping, Combining composition and task assignment into a process reduces the complexity of the problem.The experimental results show that the proposed algorithm has high quality and good balance and can effectively avoid union deadlock and resource waste.