Online tuning for retinal imaging adaptive optics systems

The imaging of the retina tissue in the back of the eye is important for a number of applications, such the early detection of eye disease. Adaptive optics systems are being increasingly used to provide clear images of the retina tissue by compensating for the eye's optical imperfections, also referred to as aberrations. Although current retinal imaging adaptive optics systems can compensate for some aberrations in the human eye, they are still unable to compensate for the unknown and time-varying higher order aberrations. This paper presents a control system design approach that allows the adaptive optics system to compensate for the unknown time-varying aberrations. The compensation for the abberations is achieved by adjusting the shape of a deformable mirror in the adaptive optics system. Hence, the control problem addressed in this paper is that of tracking an unknown and time-varying desired shape for the membrane mirror. The proposed controller design method relies on tuning, online, a Q-parameterized stabilizing controller in such a way that the tuned controller converges to the desired controller needed to achieve regulation. The resulting system will allow for improved retinal images to be taken, resulting in the early detection of eye diseases.