A system composed of a linear controller and plant can exhibit strange beha
viour when the actuators linking the controller to the plant are subject to
saturation, or other nonlinear effects. Linear conditioning means augmenti
ng the system with a linear transfer function to change the system's behavi
our during, and immediately following, saturation. In this paper, the behav
iour of linear conditioning schemes is interpreted in terms of a single tra
nsfer function M(s). This transfer function not only facilitates the unders
tanding and comparison of conditioning schemes, but also leads to a new fra
mework for conditioning design. (C) 2000 Elsevier Science Ltd. All rights r
eserved.