Our new deep loop shaping method improves control of gravitational wave observatories and helps astronomers better understand the dynamics and formation of the universe.
To help astronomers study the universe’s most powerful processes, our team has used AI to stabilize one of the most sensitive observational instruments ever built.
In a paper published today in Science, we introduce deep loop shaping, a new AI technique that opens the door to the next generation of gravitational wave science. Deep loop shaping reduces noise, improves control of observatory feedback systems, and helps stabilize components used to measure gravitational waves (tiny ripples in the fabric of space-time).
These waves are produced by phenomena such as neutron star collisions and black hole mergers. Our method will help astronomers collect critical data to understand the dynamics and formation of the universe, and better test fundamental theories in physics and cosmology.
We developed deep loop shaping in collaboration with LIGO (Laser Interferometer Gravitational-Wave Observatory) and GSSI (Gran Sasso Scientific Institute), both run by the California Institute of Technology, and demonstrated the method at an observatory in Livingston, Louisiana.
LIGO measures the nature and origin of gravitational waves with remarkable precision. But even the slightest vibration, even waves crashing on the Gulf Coast 160 miles away, can disrupt measurements. For LIGO to work, it relies on thousands of control systems that keep all its parts in near-perfect position and adapt to environmental disturbances with continuous feedback.
Deep loop shaping reduces the noise level of LIGO’s most unstable and difficult feedback loops by a factor of 30 to 100 and improves the stability of the sensitive interferometer mirrors. Applying our method to all of LIGO’s mirror control loops could allow astronomers to detect and collect data on hundreds more events per year in greater detail.
In the future, deep loop shaping could also be applied to many other engineering problems, including vibration suppression, noise cancellation, and highly dynamic or unstable systems, which are important in aerospace, robotics, and structural engineering.
