Fully distributed adaptive control for output consensus of uncertain discrete-time linear multi-agent systems

This work investigates the adaptive output consensus problem for uncertain discrete-time linear multi-agent systems on directed graphs when both Laplacian matrices for communication graphs and agent system matrices are not available. Firstly, a fully distributed algorithm is proposed to estimate the Laplacian matrix for each agent. Then, based on the proposed estimation algorithm, two fully distributed adaptive control algorithms, one for state feedback and the other for output feedback, are developed to acquire the desired controller parameters by utilizing the so-called adaptive dynamic programming techniques. It is shown that the output consensus is achieved for the resulting closed-loop multi-agent system. Simulation results demonstrate the efficacy of the proposed fully distributed adaptive controllers resulting from those adaptive control algorithms.