一种基于ISODATA算法的多智能体任务分配策略

发布时间：2018-02-28 23:33

本文关键词： 灾难救援多智能体任务分配分区策略 ISODATA算法　出处：《西北工业大学学报》2017年03期 　论文类型：期刊论文

【摘要】：针对城市地震等大规模灾害发生后,灾难救援搜救范围大、现场情况多变、远距离通讯交流阻塞、施救人员危险性高等问题,提出了一种使用具有特定功能的多智能体开展救援任务,并利用分区的思想,将多智能体分配到城市各个区域,以扩大搜救范围,节省救援时间的方案。同时,针对传统Kmeans聚类方法的不足,提出了一种基于ISODATA(迭代自组织数据分析)算法的多智能体任务分配策略,以根据不同的城市环境进行自适应聚类划分。实验结果表明,该方法不仅比传统的K-means分区算法有更好的救援效果,而且大大促进了的整体救援效果。
[Abstract]:In view of the problems of large-scale disasters such as urban earthquakes, disaster rescue and rescue, wide scope of disaster rescue and rescue, changeable situation on the spot, obstruction of long-distance communications, high risk of rescuers, and so on, In this paper, a multi-agent with specific function is used to carry out the rescue mission, and the multi-agent is allocated to every area of the city by using the idea of partition, in order to expand the scope of search and rescue and save the rescue time. At the same time, Aiming at the shortcomings of traditional Kmeans clustering method, a multi-agent task allocation strategy based on ISODATA (iterative Self-Organizing data Analysis) algorithm is proposed, which can be used for adaptive clustering classification according to different urban environments. This method not only has better rescue effect than the traditional K-means partitioning algorithm, but also greatly promotes the overall rescue effect.
【作者单位】：西北工业大学计算机学院;西北工业大学软件与微电子学院;解放军边防学院训练部教务科;
【基金】：航空科学基金(2016ZC53022) 陕西省自然科学基金(2015JM6308)资助
【分类号】：TP18;X4

【相似文献】