ADRC: a strategy to deal with Adaptive Optics feedback control

Abstract

Current advanced control strategies developed in Adaptive Optics field are based on optimal techniques in order to reject the effect of atmosphere turbulence on the phase of wavefront of the incoming light. The widely researched LQG controller relies basically on models that try to capture the temporal evolution of the atmospheric wavefront. Models based on both Cn 2 spatial priors and temporal dynamics have been proposed for SCAO and MOAO systems. The attractive performance of LQG is based on having an accurate model of the wavefront distortion. However, abrupt changes in conditions that directly affect the atmosphere produce a significant decrease in the LQG performance. To deal with this problem, an Active Disturbance Rejection Control (ADRC) strategy is proposed. ADRC, is an unconventional design strategy capable of overcoming the internal dynamics of the system and the external disturbances. The key idea is to treat the unknown information mainly due to atmosphere turbulence as state variable that can be estimated by an Extended State Observer (ESO). Instead of depending on a model, the controller draws the information needed to control the system from the ESO. Consequently, the control law for the Adaptive Optics system can be designed without the mathematical expression of the disturbances. As a result, the combined effect of the non-considered internal dynamics and the external disturbances is estimated and actively rejected. This contribution is focused on adapting ADRC method to Adaptive Optics requirements.