NN-Based Cluster Quasi-Consensus For Unknown Nonlinear Multi-Agent Systems With Input Saturations

This paper studies the cluster quasi-consensus problems for a class of unknown nonlinear multi-agent systems (MASs) with input saturations. Firstly, neural networks are used to approximate unknown functions to deal with the control difficulty caused by unknown systems. In addition, the hyperbolic tangent function is introduced to handle the influence of input saturation. Then Lyapunov theory and graph theory demonstrate the quasi-consensus of the systems. Finally, simulation results verify the effectiveness of the proposed protocols.