基于蝙蝠算法的执行器路径规划算法  

Path Planning Algorithm of Actuator in WSAN Based on BAT Algorithm

作　　者：张华 李跃飞 郑治武 ZHANG Hua;LI Yuefei;ZHENG Zhiwu(School of Computer Science and Engineering,Hunan University of Information Technology,Changsha Hunan 410004,China)

机构地区：[1]湖南信息学院计算机科学与工程学院,湖南长沙410004

出　　处：《传感技术学报》2023年第9期1445-1451,共7页Chinese Journal of Sensors and Actuators

基　　金：湖南省教育厅科学研究项目重点项目(22A0703)。

摘　　要：无线传感器与执行器网络中执行器的路径对网络能耗有着重要影响,最大化覆盖节点是收集节点数据的重要前提。为此,提出基于双重聚类的执行器路径规划(Dual Clustering-based Path Planning of Actuator,DCPA)算法。DCPA算法先利用模糊化C-均值聚类算法构建驻留点(Rendezvous Points,RPs),并依据网络失效节点动态地调整RPs位置。再利用密度峰值聚类算法将这些RPs划分成多个簇,簇数等于执行器的个数。然后,利用蝙蝠算法规划每个执行器遍历RPs的路径。考虑到蝙蝠算法的局部搜索能力的不足,加入惯性权重对蝙蝠算法进行改进。性能分析表明,相比于同类算法,DCPA算法在网络能耗和收集数据性能方面存在优势。In Wireless Sensor and Actuator Network,Path of Actuator have significant impact on energy consumption,and to maximize the number of nodes covered by actuator is critical to collecting the data from nodes.Therefore,dual clustering-based path planning of actuator(DCPA)algorithm is proposed.In DCPA algorithm,the Rendezvous Points(RPs)are selected by Fuzzy C-Means.Then,these RPs are divided into several subsets,the number of which is equal to the number of deployed actuator.Finally,the path of actuator is designed by bat algorithm,where the path covering all RPs in each subset is determined.Considering the lack of local search ability of the bat algorithm,inertia weight is added to improve the bat algorithm.Simulation analysis shows that the DCPA algorithm outperforms the existing algorithm in terms of energy consumption and collection data from nodes.

关 键 词：无线传感器执行器网络 路径规划 模糊化C-均值 密度峰值聚类算法 蝙蝠算法 

分 类 号：TN929.5[电子电信—通信与信息系统] TP212.9[电子电信—信息与通信工程] TP18[自动化与计算机技术—检测技术与自动化装置]