Speaker
Description
Modern synchrotron beamlines and neutron instruments have undergone significant changes due to technological advances and newly deployed infrastructure. Thus, experiments are becoming more data-intense and data-driven and increasingly relying on online data analysis for efficient use of experimental resources. In this regard, machine-learning (ML) based approaches of specific importance for real-time decision-making based on online data analysis and connected closed loop feedback applications.
Here we focus on a case study in x-ray reflectometry performed at ESRF using BLISS and TANGO to operate an autonomous experiment in closed-loop operation with an underlying ML model. We discuss infrastructure aspects as well as the use of ML-models in real time data analysis, essentially allowing to transfer the time spend on data analysis to a point in time prior the actual experiment.
Looking ahead, specifically in view of planed upgrade to Petra IV at DESY and the RockIT project, we also try to give some more general perspectives on the interplay of ML and autonomous experiments in beamline control environments.
Pithan et al., J. Synchrotron Rad. (2023). 30, 1064-1075
https://doi.org/10.1107/S160057752300749X
Munteanu et al., J. Appl. Cryst. (2024). 57, 456–469
https://doi.org/10.1107/S1600576724002115
| Abstract publication | I agree that the abstract will be published on the web site |
|---|
