Research on Multi-Agent Models for Collective Behavior
Jianmei Duan, Hua Li, Qiubai Sun, Xuebo Chen · 2021
Recently, the rapid development of multi-agent systems for collective behavior has become a hotspot in the field of complex networks. The paper reviews precious studies in the literatures on multi-agent models for collective behavior and classifies the models due to the different properties. For an arbitrary multi-agent model, we clear application of tasks and analyze their distribution protocols, convergence rates, and conditions for implementation. Through the simple combing of this paper, collective behavior simulation methods commonly used by scholars are introduced to the field. It is found that the commonly used behavior models can only describe most simple and orderly collective behaviors. However, the control ability for complex systems cannot be guaranteed. The optimized models can improve the simulation performance. In addition, future research directions are proposed with the aim of developing new models that focus on more robust, adaptive and efficient multiagent systems.