The purpose of this note is to consider the quadratic stabilization of LPV systems in the realistic case where only Gaussian noisy parameter measures are available. Though neglected in the actual literature on LPV systems, this question is particular important because in all situations of a practical interest the parameter measurements (or estimates) are never exact. The assumed noisy nature of physical parameter readings requires a specifically developed approach consisting of mixed robust and LPV control methods. In the present case, an approach based on a vertex result on interval time varying (ITV) matrices is proposed. This allows the solvability conditions to be stated in terms of a set of LMI's, whose number is independent of the number of time-varying parameters.
8 December 2013 in International Journal of Robust and Nonlinear Control
This technical note states new sufficient conditions for the exponential stability of linear time-varying (LTV) systems of the form aeligmacr(ldr) =A(t)aelig(t). The approach proposed derives and uses the notion of perturbed frozen time (PFT) form that can be associated to any LTV system. Exploiting the Bellman-Gronwall lemma, relaxed stability conditions are then stated in terms of ldquoaveragerdquo parameter variations. Salient features of the approach are: pointwise stability of A(ldr) is not required, ||Amacr(ldr)|| may not be bounded, the stability conditions also apply to uncertain systems. The approach is illustrated by numerical examples.
6 November 2015 in International Journal of Robust and Nonlinear Control