As the popularization of industrial intel ligence, the consensus control under switched topologies has been widely studied and applied to much practical engineering. Additionally, as the common nonlinear phenomenon, the output DZ exists in many practical systems, which will lead to the instability of system. Note that there are no publications on adaptive intelligent output feedback consensus controller for nonlinear MASs with jointly connected topologies and output DZ. Therefore, motivated by this observation, this article develops an observer-based NN consensus controller for nonlinear MASs with jointly connected topologies and output DZ by designing a distributed observer and a smooth inverse model. The developed consensus controller can guarantee the boundedness for all the variables of the controlled MASs and the convergence of the consensus errors. Moreover, the availability of the pro posed consensus control methodology is verified by the simula tion results.

This article studies the neural network (NN) adaptive consensus control issue for nonlinear multiagent systems (MASs) with dead-zone (DZ) output and jointly connected topologies. NNs are utilized to identify unknown agents, and a state observer is established to handle the problem resulted from unmeasurable states. A smooth inverse model is designed to replace nonsmooth DZ such that the nonsmooth problem of the system output is avoided. To estimate the unknown leader and its high-order derivatives under jointly connected topologies, a distributed observer is constructed. By using backstepping control technique, an output feedback consensus control scheme is established. It is demonstrated that all the variables of the controlled MASs are bounded despite of DZ output, and the followers can track the trajectory of the leader. Furthermore, multiple unmanned surface vehicles (USVs) are given to verify the feasibility of the presented consensus control methodology.