Controller Design for a High-Sampling-Rate Closed-Loop Adaptive Optics System with Piezo-Driven Deformable Mirror

Adaptive Optics (AO) systems are widely used in many scientific and medical applications, such as astronomy, laser systems and microscopes, in order to improve the resolution of the image by actively sensing and compensating the optical aberration in the system. This paper aims at improving the performance of a closed-loop AO system with Piezo-driven Deformable Mirror (PDM) and high-sampling-rate Wavefront Sensor (WFS) by means of model-based control. The improvement is achieved by reducing the hysteresis in the PDM with a hysteresis compensator and identifying a linear dynamic model of the AO system from the measurement data with a closed-loop subspace identification approach. Based on the identified model of the AO system and the model of the disturbance, a dynamic controller is designed. Experimental results show that the variance of the residual error of the proposed closed-loop AO system has been reduced by 30% with respect to the conventional AO control approach.