On a robust model-reference adaptive control system

A new type of model-reference adaptive control system is presented. The system state estimator of this adaptive system is described by the differential equation of the reference model with the feedback speeding up the estimator action during the tuning-up of the controller parameters and vanishing when the system performance approaches that of the reference model. This model-reference adaptive control system can be used to obtain the requested dynamics of the closed-loop system, which can be chosen within a wide range of speed with respect to the dynamics of the plant. Results of a simulation study show that the proposed controller can be used to control an unstable plant as well as to get a fast and well-damped response of a plant with a very small damping ratio and relatively high undamped natural frequency. The proposed approach to the estimation of the system state results in very fast estimator performance during the tuning process of the controller and in an estimation error approaching zero. This is because the estimated system performance and the estimator performance both approach that of the reference model.<
>