System Design and Control-Loop Implementation with Experimental Analysis of a High-Performance Fast Steering Mirror

Abstract

Free-space optical communication requires highly accurate fine-pointing, since small angular misalignments can degrade the optical link or interrupt it completely. Within this context, this thesis compares the system design, control-loop implementation, and experimental behavior of two coupled two-axis fast-steering mirror systems for fine-pointing, namely the commercially available FSM3000 and the newly developed miniaturized miniFSM prototype, both operated with a newly developed driver circuit. The work is carried out in a real-time MATLAB/Simulink and hardware-in-the-loop environment in collaboration with the German Aerospace Center (DLR). The driver is characterized first, followed by open-loop identification of both complete systems. Based on measured frequency responses, control-oriented low-order models are derived and used for digital controller design. For the closed-loop implementation, a PIDF-based structure in combination with the Alpha Tuning Method is employed to generate and compare controller candidates systematically. The final selection is based on a measured trade-off between closed-loop bandwidth, robustness, steady-state behavior, and the sensitivity-function shape. The results show that the same workflow can be applied consistently to both systems, that stable closed-loop operation is achieved on both axes, and that the same final controller setting is selected for both systems at α = 3 and fc = 350 Hz. The comparison further shows that the driver is the dominant hardware limitation of the present setup and that the high open-loop measurement noise prevents the desired steady-state error from being reached. Nevertheless, the implemented controller concept proves effective under random-vibration conditions, where closed-loop operation suppresses broadband disturbance significantly compared with open loop. Overall, the thesis contributes a transferable measurement-based workflow for the design and selection of digital controllers for high-bandwidth fast-steering systems under realistic hardware constraints.