Optimal regulation of stochastic linear systems with adjustable parameters

The problem of optimal regulation for stochastic linear systems with adjustable plant parameters is examined and posed as a nonlinear programming problem. A computational procedure, built around the Generalized Reduced Gradient algorithm, is developed to solve the associated plant-controller design problem. The procedure is illustrated via a lateral autopilot design in which the quality of regulation is improved by approximately 18% over that achievable with a nominal fixed plant.