The decentralized control expanding system problem (Q1106784)

Conditions are examined under which a controller for a controlled system denoted by \(S_{new}\) can be found without having to retune the previous decentralized controllers obtained in a controled system denoted by \(S_{old}.\) Given a system \(S_{new}\) in which there are no decentralized fixed modes and the robust decentralized servomechanism problem is solvable, then it has been shown that the decentralized expanding system problem is generally solvable with respect to the system's interconnection gains. Also, it is shown that a system \(S_{new}\) can be connected to an existing system \(S_{old}\) which is already satisfactorily controlled, such that for almost all interconnection gains a local controller exists for \(S_{new}\), so that the augmented composite system is satisfactorily controlled without retuning the controllers of the original system \(S_{old}.\) The problem of the paper is a generalization of the results obtained by \textit{E. J. Davison} and \textit{U. Ozguner} [ibid. 1, 255-260 (1982; Zbl 0473.93014)], which deals with the case when the control agents of \(S_{new}\) are centralized. The paper is well organized and written and can be considered to be of very good quality in its topic area.