Use of Least Means Squares Filter in Control of Optical Beam Jitter

Abstract

Abstract : Meeting optical beam jitter requirements is becoming a challenging problem for several space programs. A laser beam jitter control test bed has been developed at the Naval Postgraduate School to develop improved jitter control techniques. Several control techniques, such as least means squares and linear quadratic regulator were applied for jitter control. Enhancement in least means squares techniques to improve convergence rate was achieved by adding an adaptive bias filter to the reference signal. In the experiments, the platform is vibrated at 50 and 87 Hz. In addition, a fast steering mirror is used to inject a random component of 200 Hz band-limited white noise. The experimental results demonstrated that the addition of the adaptive bias filter to the least means squares algorithm significantly increased the converging rate of the controller. To achieve the reduction of both sinusoidal and random jitter, a combination of least means squares/adaptive bias filter and linear quadratic regulator is most effective. The least means squares/adaptive bias filter control is most effective for a sinusoidal jitter and the linear quadratic regulator control for a random jitter.