Reinforcement learning is finding its way to real-world problem application, transferring from simulated environments to physical setups. In this work, we implement vision-based alignment of an optical Mach-Zehnder interferometer with a confocal telescope in one arm, which controls the diameter and divergence of the corresponding beam. We use a continuous action space; exponential scaling enables us to handle actions within a range of over two orders of magnitude. Our agent trains only in a simulated environment with domain randomizations. In an experimental evaluation, the agent significantly outperforms an existing solution and a human expert.
@article{arxiv.2107.04457,
title = {Aligning an optical interferometer with beam divergence control and continuous action space},
author = {Stepan Makarenko and Dmitry Sorokin and Alexander Ulanov and A. I. Lvovsky},
journal= {arXiv preprint arXiv:2107.04457},
year = {2021}
}