| Abstract: | In this paper the parameter estimation problem for deterministic MIMO systems with overparametrized models will be addressed. In SISO systems overparametrized signal models might arise if the order of the plant model is set too high. For MIMO systems this problem will arise whenever its observability indices are different or, as in SISO systems, the order of the system is set too high. With a proper definition of persistent excitation it is shown that the estimated parameters will converge to a set of parameters. Each point of this set will result in the same transfer function as that of the system under consideration. A very efficient correction algorithm will be used to remove the greatest left common divisor of the estimated system parameters. Hence adaptive control algorithms, which may not be suitable if the estimated system parameters are not left coprime, can then be implemented. |
