Applied Soft Computing ( IF 8.7 ) Pub Date : 2020-08-27 , DOI: 10.1016/j.asoc.2020.106680 Daifeng Zhang , Jiliang Zhang
The coverage optimization of wireless visual sensor networks (WVSNs) with changeable field of views (FOVs) brings more challenges on the decision making of modern cyber–physical systems as it is a multi-objective decision problem where some contradictory indices, e.g. the sensor coverage and the redundancy, should be taken into consideration concurrently. In this paper, a novel multi-species evolutionary algorithm (MSEA) is proposed to address this issue by introducing a multi-species evolution scheme to enhance the search ability. A competition mechanism based on the deductive sort and the crowding distances is developed to facilitate the generation of Pareto front and the elitist individuals are evolved from a multi-species hybrid population. Comparative results show a better balancing performance between the exploration and the exploitation of the proposed algorithm which induces a strong approximation to the feasible WVSN managements.
中文翻译:
视野可变的无线视觉传感器网络覆盖优化的多物种进化算法
具有可变视场(FOV)的无线视觉传感器网络(WVSN)的覆盖范围优化对现代网络物理系统的决策提出了更多挑战,因为它是一个多目标决策问题,其中一些矛盾的指标(例如传感器)覆盖范围和冗余度应同时考虑。本文提出了一种新颖的多物种进化算法(MSEA),通过引入一种多物种进化方案来增强搜索能力。建立了一种基于演绎类别和拥挤距离的竞争机制,以促进帕累托阵线的产生,而精英分子则是从多物种混合种群中进化而来的。