Search

This note proposes a dynamic controller for practical coordinated tracking of uncertain heterogeneous multi-agent systems. The agents are high-order linear systems subject to external disturbances and plant uncertainties, the input to the leader is not known to other agents, and the network topology is time-varying. We adopt a recently developed reduced-order disturbance observer to present a distributed tracking controller which uses only relative measurements. It is shown that the proposed controller can be tuned to ensure the ultimate boundedness of tracking error which can be chosen arbitrarily small despite the network switching, plant uncertainties, and disturbances. Simulation results are provided to show the effectiveness of the proposed controller. [ABSTRACT FROM PUBLISHER]

The disturbance observer has been widely employed in applications due to its powerful ability for disturbance rejection and robustness under plant uncertainties. However, it rejects the disturbance approximately rather than exactly since it is usually designed without considering structural properties of disturbance. In order to improve the disturbance rejection performance, we propose a design method to embed the internal model of disturbance into the disturbance observer structure. Furthermore, a systematic design procedure is proposed so that one can always design the disturbance observer to guarantee robust stability of the closed-loop system even though uncertain parameters of the plant belong to an arbitrarily large (but bounded) set. [ABSTRACT FROM AUTHOR]

An output-feedback controller has been recently proposed that has the following features: 1) it is an inner-loop controller so that it can be added on the existing closed-loop system working in harmony with a pre-designed (possibly non-robust) outer-loop controller; 2) it robustifies the closed-loop system in a way that the uncertain plant under external disturbance behaves like a disturbance-free nominal plant; and 3) it recovers the trajectory of the nominal closed-loop system in time domain. However, it is restricted to the single-input-single-output nonlinear systems. In this technical note, we extend this result for a class of multi-input-multi-output (MIMO) nonlinear systems having the same number of inputs and outputs. The tools used in this synthesis are the singular perturbation theory and the multi-variable circle criterion. An example shows the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]