抄録
In this paper, the controllability of decentralized control systems is considered. Stations of a decentralized control system can use their control inputs not only to control the system actually, but also to exchange their information one another through the state space. Therefore it is important to take account of such co-operation mode explicitly when we consider the controllability of decentralized control systems.
The controllability is defined as follows in this paper: A decentralized control system is controllable if an arbitrary initial state can be transferred to the origin of the state space in a finite time interval by the stations with decentralized control laws.
First of all, a pseudo-order relation is introduced among the stations, and the largest observable subspace that each station can obtain with the above-mentioned co-operation mode is given explicitly. Then the largest reachable observable subspace that each station can control by co-operation with other stations is given. Based upon these results, a necessary and sufficient condition of the controllability is derived. The condition is formulated with Hanlek matrices.
