基于智能体异质影响的观点一致性控制研究

Research and Control on Opinion Consistency Based on Heterogeneous Influence of Agents

针对传统观点模型中个体影响能力相同的简单设定,本文基于社会网络结构的拓扑属性,将节点度相关的时变指数权重引入连续观点模型,提出异质权重观点演化动力学模型,研究多智能体系统中的个体异质性对观点演化行为的影响,并从网络拓扑结构角度对观点一致性进行分析和控制.仿真结果表明与传统模型相比,网络个体的异质性相互作用极大地改变了观点演化状态,促进了整个系统的观点一致性达成.随着权重指数增大,个体的异质性影响能力增加,使网络连通性增强.网络中节点度较大的个体的主导作用促使观点能够更快地收敛达成一致,从而提高了系统收敛效率.与固定权重观点模型相比,权重随观点的动态演化使系统能够在相对有限的交互邻域内更快地收敛.

Aiming at the simple setting of the same individual influence ability in the traditional opinion models,based on the topological properties of social network structure,this paper introduces the time-varying exponential weight related to the node degree into the continuous opinion model and proposes a heterogeneous weight opinion evolution dynamics model.The influence of individual heterogeneity in multi-agent systems on opinion evolution behavior and opinion consistency are analyzed and controlled from the perspective of network topology.The simulation results show that,compared with the traditional model,the heterogeneous interaction of network individuals greatly changes the evolution state of opinions and promotes consensus of the whole system.With increasing of the weight exponent,the individuals’heterogeneous influence ability increases,which makes the network have better connectivity.The dominant role of individuals with relatively large node degrees in the network promotes faster convergence of opinions to consensus,thereby improving the convergence efficiency of the system.Compared to opinion models with fixed weights,the dynamic evolution of weights with opinions enables the system to converge faster within a relatively limited interaction neighborhood.