Hybrid adaptive compensation for servo disturbances harmonic in displacement

The authors consider servo systems with geared transmissions, which can experience significant disturbance forces that are periodic in displacement. They assume known spatial frequencies and use this knowledge to reduce the number of parameters required for estimating the disturbance. Use of this partial knowledge of the disturbance is expedited by a continuous-time disturbance model implemented in a sampled data form (hybrid estimation). The estimated disturbance is then used to generate a feedforward compensation signal that is added to a feedback control signal to improve system performance. A difficulty encountered in performing the parameter estimation is the tendency of the parameters to vary with time given simple harmonic inputs. This tendency can be reduced by averaging but sensitivity to parameter variation is lost.