Search

This paper presents a data-driven algorithm for tracking control of a linear discrete-time plant. No traditional mathematical model of the plant such as transfer function or state equation is employed. Instead, the plant dynamics is represented by an array whose elements are plant input-output data. Sensitivity of the array to input and output noise is introduced. Then, it is shown that, for an arbitrary reference signal, the tracking control input which minimizes the sensitivity can readily be computed by solving a linear matrix inequality which is composed of the array.

Robust control systems synthesis is generally recast as a class of robust feasibility problems which is to find a solution satisfying a set of parameter-dependent convex constraints for all possible parameter values. For this class of the problems, a stochastic ellipsoid method with multiple cuts each of which corresponds to each of the constraint is proposed, where a new update rule is presented for constructing a smaller ellipsoid which contains the intersection of a previous ellipsoid and half spaces determined by given multiple subgradients. Moreover, we show an explicit relation between the volume of the ellipsoid updated by the original method and that of the proposed method. A quantitative analysis of the volume of the updated ellipsoid is also provided, which leads to a further modification of the algorithm for achieving fast convergence.

Optimal tracking is considered for a linear time-invariant plant in data space. The proposed control strategy does not employ any traditional mathematical model such as a transfer function or a state-space equation. Instead, the plant dynamics is represented as a set of basis vectors whose elements are input-output data of the plant. Using this system representation, an optimal tracking problem is solved, which is to find the control input which minimizes a quadratic performance index subject to achieving dead-beat tracking for arbitrary reference signals.

Optimal control is considered for a linear time-invariant plant in the input-output data space. The proposed control strategy does not employ any traditional mathematical model such as a transfer function or a state-space equation. Instead, the plant dynamics is represented as a set of basis vectors whose elements are input-output data of the plant. Using this system representation, two optimal control problems are solved. One is to find the control input which minimizes a quadratic performance index. The other is to find the control input which minimizes a quadratic performance index subject to achieving dead-beat tracking.

In this paper, connective stability is considered for large scale systems consisting of a number of subsystems and uncertain interconnections. A necessary and sufficient condition of connective stability is derived in terms of linear matrix inequalities (LMIs), under which the overall systems are robustly stable against the uncertainties of the interconnections. This condition gives a useful sufficient condition which lead to an LMI condition even for connective stabilization via decentralized state feedback if an unstructured matrix variable in the LMIs is restricted to a block diagonal one. In this paper, a characterization of this sufficient condition is also derived in frequency domain. It is shown that the condition is equivalent to a familiar ℳ matrix condition which utilizes ℋ∞ norm of each subsystem.

This paper considers robust decentralized control of large space structures composed of interconnected substructures. A local controller is applied to each substructure with collocated displacement/angle sensors and force/torque actuators. The idea of DVDFB (direct velocity and displacement feedback) is adopted for control robustness. A filter of a relative degree 2 is introduced at each input channel to approximate DVDFB by a strictly proper controller for implementation. It is shown that the overall closed-loop system becomes stable and optimal for a frequency-dependent quadratic cost function by choosing the filter parameters and the DVDFB gains appropriately.

This paper considers position and attitude control of large space structures composed of a number of subsystems which are interconnected by springs and dampers. It is assumed that sensors and actuators are collocated. A decentralized simple dynamic displacement feedback, proposed by Fujisaki, Ikeda, and Miki, is transformed to a controller whose input contains the control input and the measured output of the space structure. The objective of this paper is to present a condition under which the controller is optimal for a quadratic cost function.

This paper considers a decentralized H"~ control problem for multi-channel linear time-invariant systems. Our interest is focused on dynamic output feedback. The control problem is reduced to a feasibility problem of a bilinear matrix inequality (BMI). The objective of this paper is to propose an algorithm for solving the BMI by using the idea of the homotopy method, where the controller's coefficient matrices are deformed from full matrices defined by a centralized H"~ controller, to block-diagonal matrices of specified dimensions which describe a decentralized H"~ controller. When a feasible decentralized H"~ control problem is considered, it can be expected that there always exists a centralized H"~ controller for which the algorithm converges and presents a desired solution. To find such a suitable centralized H"~ controller, random search in a parametrized set of H"~ controllers with a proper dimension is suggested. The efficiency of the proposed algorithm is demonstrated by an example.

This paper considers the general mixed H 2 / H ∞ control problem with no assumption on the plant matrices. It is the problem of finding an internally stabilizing controller which satisfies two inequality constraints on both the mixed H 2 / H ∞ performance measure and the H ∞ norm of another closed-loop transfer function. Since the performance measure is an upper bound on an H 2 norm of a closed-loop transfer function, the transfer function must be strictly proper to satisfy the corresponding inequality constraint. This implies that admissible controllers must be strictly proper in case no assumption is made on the plant matrices. In this paper, necessary and sufficient conditions for the existence of a strictly proper H 2 / H ∞ controller are given in terms of linear matrix inequalities, and an adequate strictly proper controller whose order equals to the plant order is given explicitly. The solvability conditions for pure H 2 and H ∞ control problems with strictly proper controllers are also derived.

In this paper, we deal with modeling and controller design of coke oven batteries. Mathematical models on heat balance of carbonization chambers and combustion chambers are introduced and their combination gives the thermal dynamics of the whole battery systems. A cycle of processes in carbonization chambers is an endothermic chemical reaction and a post-carbonization, which causes thermal oscillation, and we give a model on actual charging schedule of coal to carbonization chambers, called ``5 skip operation.'' In order to restrain the temperature oscillations in chambers under parameter uncertainties, we propose a decentralized adaptive controller and show that it attains convergence to ideal time responses.