LQG feedforward/feedback stochastic optimal control and marine application

摘要

The solution of a 3.5 degree-of-freedom H{sub}2 stochastic optimal control problem is discussed. The system is assumed to be scalar and includes a 2.5 degree-of-freedom tracking controller, together with a feedforward control action. Theoptimisation is based upon an extended cost-function which allows separate costing of the terms due to the feedback, reference and feedforward controllers. This enables the amount of feedforward action to be varied relative to the feedback action and therobustness of the feedback-loop to be shaped independently of the tracking performance. The use of a separate cost term to be able to vary the amount of feedforward action and the resulting design steps are novel and important in applications. Previoussolutions did not allow separate cost tuning of the feedforward control terms. A ship positioning design study, that utilises feedforward action, is presented to illustrate the benefits of the approach.