Automated Spectrometer Alignment via Machine Learning

What is it about?

Applying the use of neural networks and machine learning methods to assist scientific experiments at leading research facilities is becoming more widespread. We have demonstrated an AI method for the automated alignment of a Spectrometer at a synchrotron light source. This method uses a neural network trained with simulated spectroscopy data and employs an optimisation loop to overcome the offset between synthetic data and experiment data.

Featured Image

Photo by Google DeepMind on Unsplash

Photo by Google DeepMind on Unsplash

Why is it important?

Due to the cost of beam time at synchrotron light sources time spent setting up and running experiments needs to be reduced. Our method of automatic alignment of components of a spectrometer aims to accelerate a traditionally time consuming task.