Abstract

Adaptive optical techniques involve the closed loop processes of wavefront error sensing and corrective control based on some performance criteria dictated by the intended application. Error sensing and corrective control procedures employed both in analytical and experimental adaptive optical studies to date involve generally direct techniques and fixed performance criteria. That is, current procedures employ wavefront compensatation control that is directly related to sensed errors (as in the case of interferometric phase measurement type systems) or directly derivable from sensed signal modulations (as in the case of amplitude modulated multidither type systems). Further, the controlling performance criteria are generally fixed. These approaches exclude considerations of the dynamic nature of the wavefront distortion processes which the system is attempting to compensate. Also, corrective trends successfully executed by the control process are neither identified nor employed during the continuously cycled corrective loop. This study will assess the merits of adding to the conventional adaptive optical convergence algorithm the sophistication of estimation and adaptive performance criteria. These features will be configured to sense and in real time predict the behavior and dynamic processes involved in wavefront distortion for the purposes of improved compensation performance, convergence responses, and stable criteria optimization. Details will be given regarding the behavior of convergence and error sensing algorithms that are based on estimation and predictive filtering theory. Figures of merit will be defined and examined to performance compare these adaptive algorithm approaches with conventional approaches. The degree of increased sophistication, modeling, and hardware implementation for these approaches will also be discussed and compared with conventional approaches.

© 1976 Optical Society of America