基于拍卖的动态任务分配算法

摘要

Dynamic task allocation is a key issue of open multi-agent system(MAS). Based on the algorithm given by the reference [9] , this paper presents an improved algorithm for a new task appearing at any time in the process of task allocation or executing. The paper takes into account the Agent's expectation of the known tasks and future tasks, in order to guide the agent reasonable bid to get the optimum alloction. With multi-agent transporting army supplies for background, the results show that the algorithm is reasonable and effective.%动态任务分配是开放式MAS系统研究的重点.在文献[9]给出的算法基础上,针对任务分配或执行过程中,随时有新任务出现的情况,本文提出一种改进算法,同时考虑Agent对已知任务和未来任务的期望度,以引导Agent合理出价,从而得到较优的分配结果.通过在多Agent运输军需问题中的应用,验证了该算法是合理有效的.