Abstract: An adaptive robust controller based on dynamic filtering method is presented for the wide vari- ations in rotational inertia and loading moment and the input control constraints of a rocket launcher posi- tion servo system. A projection-type estimation algorithm is used to estimate the uncertainties of model, the design of the controller is simplified by using a robust sliding model filter, meanwhile an auxiliary a- nalysis system is utilized to handle a saturation problem, Besides, the auxiliary analysis system is com- bined with the backstepping method, a design method of controller is presented by equivalently transfor- ming the model and selecting a proper Lyapunov function. The simulation results show that the proposed approach can guarantee the response speed and control accuracy, and possess a strong robustness to both parameter perturbation and load disturbance.

Key words: automatic control technology, antiaircraft rocket launcher, input control constraints, dynamic surface filtering method, robust adaptive backstepping