Adaptive H∞ control for linear slider with friction compensation - using gradient algorithms with projection -

In this paper an adaptive tracking control of a servo mechanism with friction is developed, based on the notion of H^infin optimality. It is assumed that the friction dynamics can be described by the LuGre model. The nonlinear characteristic function of the LuGre model is parameterized by a neural-network (NN). In general, friction parameters and NN weight are unknown. These parameters are updated by estimating strategy in this proposed method. An approximation error in NN is regarded as an exogenous disturbance to the system, the L₂ gains from the disturbance to generalized outputs are made less than prescribed positive constants. To cope with the practical applications, gradient algorithms with projection method is also applied to the unknown parameter estimation methods. The effectiveness of the proposed method is evaluated via some experimental results