Adaptive Control of Linear Time-Varying Plants: A New Controller Structure

Abstract

In this paper we consider the Model Reference Control and Model Reference Adaptive Control problem of a class of linear time-varying plants with bounded, piecewise smooth parameters. We first introduce a new controller structure that solves the Model Reference Control problem, with zero tracking error, in the case of smooth slowly time-varying plant parameters. When the plant parameters are unknown we use the robust adaptive law, proposed in [12], to adjust the controller parameters and establish boundedness of all signals in the adaptive loop and good tracking properties, provided that the smooth time-variations of the plant parameters are sufficiently slow and the minimum time between discontinuities is large enough. Furthermore, we show that when the plant parameters become constant, the tracking error converges asymptotically to zero.