Intelligent use of prior information in adaptive control design

In this paper, we develop simple adaptive controllers which make intelligent usage of prior information. These controllers are low-order, have few adjustable parameters, and hence are easy to implement. By utilizing the structure of the plant to be controlled, there are several instances where the adaptation can occur in a simplified form. We discuss the conditions under which such simple controllers can be realized and establish their stability. Their robustness properties in the presence of measurement disturbances, time-varying parameters, and unmodeled dynamics is discussed. The behavior of the resulting controllers is evaluated both experimentally and through simulation studies of complex dynamic models of different applications.