New solution to adaptive inverse control

Abstract

The inverse control is a solution to the adaptive control problem, in which the controller approximates the inverse of the plant model. In other words, the controller transforms the input sequence to a sequence that when applied to the plant input, the plant output will be the same as the input sequence, or at least the same as its delayed replica. In this paper the previous works on the subject are extended by introducing a nonadaptive feedback from the plant output to the controller input. It is shown that this extension results in more robust implementation. The stability of the proposed scheme is also addressed and solutions are given to overcome the instability problems that may arise under some rare conditions.