A decentralized adaptive control scheme is presented for the stabilization of interconnected subsystems. The proposing decentralized control scheme is based on the conditions that the information required for the design of each subsystem and the measurement available for each control are restricted to those of that subsystem. The magnitude of the interconnection and/or the information about the other subsystems are not necessary for the design and the control.
The assumptions about the controlled plant are the followings; (A1) the interconnection enters into each subsystem through its input, and (A2) the transfer function from the input to the output of each subsystem is minimum-phase and its relative degree is one. The assumption (A1) is made for, decentralized stabilizability of the overall system, while (A2) for robust high gain stabilizability of each subsystem.
The proposing control law consists of the output feedback of each subsystem, and each feedback gain is tuned adaptively. The σ-modification is used in the parameter adjustment law since the control objective here is small regulation errors with small feedback gains. The stability analysis shows that the solutions of the system are bounded where disturbances are bounded.
Numerical analysis is performed to illustrate the robustness of this method to the bounded disturbances and the controller breakdown in one of subsystems.