A Kalman Filter Optimization Approach to Direct Adaptive Control

This paper presents a novel Kalman filter based approach for approximately enforcing a linear constraint in adaptive control. One application is that this leads to an alternative forms for well known modification terms such as emodification. It is shown that employing this approach does not increase the theoretical guaranteed ultimate bounds for the closed loop error signals of an existing adaptive design. In addition, it leads to smaller tracking errors without incurring significant oscillations in the system response, and without requiring high modification gain that can excite the unmodeled dynamics. Three novel modifications to classical adaptive laws are illustrated and tested on a model of wing rock dynamics.