Bayes Control of Hammerstein Systems

In this paper, we consider data driven control of Hammerstein systems. For such systems a common structure is transfer function followed by static output nonlinearity that tries to cancel the input system, which modeled as polynomial or piece-wise linear function. The part controller used achieve desired disturbance rejection and tracking properties. To design controller, propose weighted average risk criterion with being squared L2 error. Here respect observations in weighting how important it have good for different impulse responses. This corresponds leading Bayes estimator therefore call approach control. By parametrizing estimating corresponding hyperparameters tune information regarding true response contained set available user design. numerical results show proposed methods result stable controllers performance comparable optimal designed using plant.