Improving cosmological reach of a gravitational wave observatory using Deep Loop Shaping
This addresses the critical bottleneck of control noise for improving sensitivity in gravitational wave observatories, enabling better detection of astrophysical events like black hole mergers.
The paper tackled the problem of harmful noise injected by mirror stabilization control in gravitational wave observatories, which limits low-frequency sensitivity, and demonstrated that their Deep Loop Shaping reinforcement learning method reduced control noise by over 30x in the 10-30Hz band and up to 100x in sub-bands at the LIGO Livingston Observatory.
Improved low-frequency sensitivity of gravitational wave observatories would unlock study of intermediate-mass black hole mergers, binary black hole eccentricity, and provide early warnings for multi-messenger observations of binary neutron star mergers. Today's mirror stabilization control injects harmful noise, constituting a major obstacle to sensitivity improvements. We eliminated this noise through Deep Loop Shaping, a reinforcement learning method using frequency domain rewards. We proved our methodology on the LIGO Livingston Observatory (LLO). Our controller reduced control noise in the 10--30Hz band by over 30x, and up to 100x in sub-bands surpassing the design goal motivated by the quantum limit. These results highlight the potential of Deep Loop Shaping to improve current and future GW observatories, and more broadly instrumentation and control systems.