Data-driven subspace predictive control: lab demonstration and future outlook

Abstract

The search for exoplanets is pushing adaptive optics systems on ground-based telescopes to their limits. A major limitation is the temporal error of the adaptive optics systems. The temporal error can be reduced with predictive control. We use a linear data-driven integral predictive controller that learns while running in closed-loop. This is a new algorithm that has recently been developed. The controller is tested in the lab with MagAO-X under various conditions, where we gain several orders of magnitude in contrast compared to a classic integrator. We will present the lab results, and we will show how this controller can be implemented with current hardware for future extremely large telescopes.

Files

Additional details