分布式约束优化问题及其求解算法

分布式约束优化问题(DCOP)能够对多智能体系统(MAS)中的各种分布式推理任务进行建模,广泛应用于分布式规划、调度、资源分配等问题中。首先从DCOP的概念出发,引入一个典型的DCOP实例,在此基础上对DCOP问题求解的两类主流算法进行了详细介绍和比较分析。针对DCOP对现实问题建模中出现的部分集中式、硬约束、开放式、隐私和anytime等5个方面的问题进行了阐述,并介绍了相应的扩展算法。在动态实时问题,自稳定性与误差容错以及在物理分布式环境下仿真等问题仍需进一步研究。

Research on Distributed Constraint Optimization Problem and Solutions

The Distributed Constraint Optimization Problem(DCOP) is able to model a variety of distributed reasoning tasks that arise in multiagent systems, and is widely applied distribute programming,scheduling and resource allocation etc.Based on concept of DCOP and a typical example,this paper analyses two main types of algorithms on solving DCOP.To solve five problems existing in modeling real-world with DCOP.Such as:Partial Centralization,Hard Constraints,Open,Secret,anytime problems etc,the further expounding of these problems and the extension of the corresponding algorithms are introduced.Further study is stil necesseny in the dynamic real-time,Self-stabilizing and Fault-containing as well as simulating in physical distributed environment and other issues.