MODEL-BASED CONTROL OF THE BURGERS-EQUATION

We extend a method of control, model-base control, to the realm of partial differential equations. The hallmark of model-based control is that a particular goal dynamics is achieved by using a model for the observed dynamics to create the appropriate driving needed to make the goal dynamic an attractor for the system, alleviating the need for constant feedback, as is necessary with traditional control methods. We investigate a model-based control for the Burgers equation, with particular attention to sensitivity of the control to model inaccuracies, boundary errors, and noise.