基于独立学习的多智能体协作决策  

Multiagent cooperative decision making based on independent learning

作　　者：李晓萌[1] 杨煜普[1] 许晓鸣[1] 

机构地区：[1]上海交通大学自动化研究所,上海200030

出　　处：《控制与决策》2002年第1期29-32,共4页Control and Decision

摘　　要：联合学习模式是实现多智能体协作决策的有效方法 ,但是当智能体信息不完备时 ,这一方法难以适用。为此 ,在智能体独立学习的基础上提出一种多智能体协作决策方法。以网格对策为例 。Although joint learning is an efficient method to implement multiagent cooperative decision, it is unsuccessful when agent has imperfect information. The method of agents′ independent learning which acts as the base of multiagent cooperative decision is put forward. The experiment of grid games shows the efficiency.

关 键 词：独立学习 多智能体协作决策 智能控制 学习算法 联合学习模式 

分 类 号：TP293.5[自动化与计算机技术—检测技术与自动化装置] TP18[自动化与计算机技术—控制科学与工程]