Analysis of DKIST Adaptive Optics Jitter Control

The Daniel K. Inouye Solar Telescope (DKIST) is a 4-meter solar telescope currently under construction at Haleakala, Hawaii. Its science objectives include studying the dynamics of the outer solar atmosphere and magnetic structures of the Sun. In order to obtain high resolution images for scientific study, DKIST requires an accurate adaptive optics (AO) system that will correct atmospheric aberration of incoming wavefronts. By analyzing the jitter control system, which corrects image shift, we can identify and eventually eliminate sources of non-atmospheric vibration that cause jitter and reduce image resolution. In the present work, we constructed movies and used image cross-correlation techniques to corroborate jitter seen in images with jitter calculated using Python code. After quantifying the magnitude of jitter in images, we used power spectra of AO mirror command telemetry data to determine the strength of jitter vibrations at different frequencies. We also compared the response of the actual jitter control system to the response of a Matlab model of the system. Analysis indicates that the system correction is matched by the model in the 5 Hz to 100 Hz range, suggesting that the atmospheric simulation matches atmospheric data at lower frequencies. But we also observe continuum mid-frequency (40 - 150 Hz) jitter unexplained by atmospheric or telescope wind-shake modeling. By fitting the modelled system correction to the actual system response, we can identify input parameters, such as integral and proportional gain, that need to be adjusted to obtain a better correction. We present methods used to quantify the AO jitter performance and discuss ways to further mitigate jitter through mechanical damping or more advanced control algorithms.