LPV-ARX representations of LPV state-space models with affine dependence (Q6103082)

In this paper, the authors prove that an input-output map can be given by a linear parameter-varying (LPV) state-space representation with an affine and static dependence on the scheduling variables, if and only if this input-output map satisfies certain LPV autoregressive input-output equations. The LPV autoregressive input-output equations is linear in the derivatives (for continuous-time) or time-shifts (for discrete-time) of the outputs and control inputs, while the coefficients of this linear equation are polynomials of the shifts of the scheduling variable in discrete-time, or of the high-order derivatives of the scheduling variable in continuous-time. In some sense this result is a generalization of the well-known equivalence between linear state-space representations and autoregressive input-output models, and this result extends a well-known 12 years old results on LPV state-space representations with a dynamic and meromorphic dependence on the scheduling variables to LPV state-space representations with a static and affine dependence on the scheduling variables.