Actuator requirements in adaptive control systems

Technological processes invariably become more complex while expectations on quality and efficiency increase. As an example, consider the problem of controlling the speed of a drive system, which is part of a mechatronic process. A typical requirement is to track references of torque and speed fast and accurately. At the same time, the system should be able to handle uncertainties arising from unknown parameters, disturbances as well as nonlinear effects. In order to meet such requirements an increasing amount of sophisticated control theory is put into the system. It is often only at second glance that one realizes that the use of new controllers has important consequences concerning the performance of the actuator subsystem. In standard cascaded PI(D)-control the nature of the control input can be predicted either from experience or by invoking some classical control theory. The choice of an actuator, which generates the input, is then a straightforward task. This ceases to be the case when more complex control strategies are used. In most cases, analysis is difficult due to the nonlinear nature of the closed loop system. The objective of this paper is to provide a study of the actuator requirements that can be expected when the controller is adaptive.