Hierarchical Decentralized Robust Optimal Design for Homogeneous Linear Multi-Agent Systems

This paper proposes novel approaches to design hierarchical decentralized robust controllers for homogeneous linear multi-agent systems (MASs) perturbed by disturbances/noise. Firstly, based on LQR method, we present a systematic procedure to design hierarchical decentralized optimal stabilizing controllers for MASs without disturbances/noise. Next, a method for deriving reduced-order hierarchical decentralized stabilizing controllers is presented by suitable selections of the weighting matrices in the LQR performance index. Secondly, the hierarchical decentralized robust controller designs in terms of $H_{\infty}$ and $H_{2}$ norms are introduced, which include two different scenarios namely general and LQR-based synthesis. For the general synthesis, the robust controller gains are computed as solutions of a distributed convex optimization problem with LMI constraints. On the other hand, for the LQR-based design, the robust controller gains obtained from the general synthesis are further verified as LQR stabilizing gains to be unified with the LQR-based design when there are no disturbances/noise. This results in a hierarchical decentralized inverse optimal control problem, for which we will propose a new method to resolve it. Finally, several numerical examples are presented to illustrate the effectiveness of the proposed approaches.