This paper describes a new adaptive controller for MIMO systems. The resulting closed-loop system will be stable and decoupled. The system to be controlled can be linear, non-minimum phase and/or unstable. The interactor matrix need not be known beforehand. Instead of employing the certainty equivalence principle to design the adaptive controller, a special controller design procedure is used. By using this special scheme, the computation time required for controller design will be drastically reduced. Moreover, the boundedness of the controller parameters is automatically ensured. This is a desired property which is useful in stability analysis. The steady state performance of the controller designed by the proposed method will be the same as that of controllers designed using the certainty equivalence principle. Approximate decoupling can be achieved by underestimating the degree of the decoupling matrix without affecting the stability of the system. To achieve decoupling, we need to specify the denominator of the closed-loop transfer function matrix of the system which is assumed to be in the form T(q−1) = t(q−1)I where the polynomial t(q−1) is assumed to be monic and stable. Global stability of the adaptive system is obtained for deterministic systems. A simulation example is presented to demonstrate this control scheme.
[1]
Michel Kinnaert,et al.
A new decoupling precompensator for indirect adaptive control of multivariable linear systems
,
1987
.
[2]
Duncan A. Mellichamp,et al.
A decoupling pole placement self‐tuning controller for a class of multivariable processes
,
1986
.
[3]
G. Goodwin,et al.
Adaptive control of nonminimum phase systems
,
1981
.
[4]
I. D. Landau,et al.
Quasi-Direct Adaptive Control for Nonminimum Phase Systems
,
1982
.
[5]
L. Mo,et al.
Consistent parameter estimation in adaptive control for MIMO systems
,
1989
.
[6]
W. Wolovich,et al.
Arbitrary adaptive pole placement for linear multivariable systems
,
1984,
1982 21st IEEE Conference on Decision and Control.
[7]
P. Ramadge,et al.
Discrete Time Stochastic Adaptive Control
,
1981
.
[8]
Rolf Johansson,et al.
Parametric models of linear multivariable systems for adaptive control
,
1982,
1982 21st IEEE Conference on Decision and Control.