Consensus output regulation without state estimation for a class of nonlinear systems

This paper deals with consensus output regulation of network connected multi-agent systems. Every agent or subsystem is a nonlinear system in the output feedback form, and has identical dynamics. The subsystem dynamics are influenced by state variables generated from an exosystem. The outputs of the subsystems are required to follow a desired trajectory which is a function of the exosystem state. Only some subsystems have access to the desired trajectory, and other subsystems will have to rely on the exchange of information through the network. In this paper, a consensus control design is proposed to ensure that the outputs of all the subsystems converge to the same desired output trajectory by exploiting the internal model design strategy. The proposed control design only uses the relative outputs of the subsystems, and does not require the estimation of subsystem state variables.