摘要
任务分解是多Agent系统中合作问题求解的重要步骤之一。常用任务分解方法大多是基于任务本身的特征进行的,较少考虑任务执行体Agent的特征,这样可能导致分解的任务不能实现或带来过多的协调费用。针对该问题,本文利用基于Agent能力的形式化方法对任务分解问题进行描述,将该问题转化为可行操作集的求解问题。设计了一种基于启发式算法的分解策略,并对该算法进行了性能分析和示例。
Task decomposition is one of the important step of cooperative problem-solving in the Multi-Agent System.The common methods mostly take the characteristic of tasks into account without the ability of Agent involved.It will make task decomposition implemented difficultly and bring overfull costs.In order to solving this problem,the paper characterizes the problem of task decomposition by the formal specification method based on the ability of Agent.The problem is transformed to the problem that how to acquire feasible set of operations.Then,a strategy of task decomposition based on heuristics algorithm is proposed.Finally,performance of the algorithm is analyzed and a demonstration is gave.
出处
《煤炭技术》
CAS
北大核心
2010年第12期201-203,共3页
Coal Technology
关键词
多AGENT系统
任务分解
可行操作集
启发式算法
Multi-Agent System
task decomposition
feasible set of operations
heuristics algorithm
