Estimation of Unknown Disturbances in Gimbal Systems

摘要

This paper presents a method for the estimation of unknown disturbances for high precision gimbal systems. Alternative to the classical methods of model inversion and filtering, we employ an asymptotically stabilizing controller which achieves the estimation process even in the presence of unstable zeros. The architecture provides an input equivalent disturbance which can be thought to capture all real disturbances in the system as well as virtual ones such as unmodelled dynamics and nonlinearities. The proposed method is illustrated on a 2-axis gimbal system where system identification is followed by the design on an integral linear quadratic regulator as the stabilizing controller which forms the base of the disturbance observer. It is seen that a random unknown disturbance is estimated successfully in the presence of additional gyro noise.