AN ADAPTIVE DIRECT DATA DRIVEN CONTROL SCHEME FOR UNKNOWN PLANT

This paper develops an adaptive direct data driven control scheme for unknown plant in one closed loop system, whose goal is to achieve plant-model perfect matching condition. To apply direct data driven control to designing forward controller without the model of plant, virtual input is constructed to derive one optimization problem, whose decision variables are the unknown controller parameters. Through using the idea of adaptation, one parameter adjustment loop is added as the outer loop in such a way the unknown controller parameters are changed with environment, according to our constructed parameter adjustment law. Furthermore Lyapunov's stability theory is also used to derive parameter adjustment law such that stability can be guaranteed for the whole adaptive system. Such an adaptive direct data driven control can not only design controller without the model of plant, but also adjust unknown controller parameters adaptively through one constructed parameter adjustment mechanism. Finally two simulation examples confirm our theoretical results.