TomOpt: Differentiable Muon-Tomography Optimization
- 1Research Institute in Mathematics and Physics, Université Catholique de Louvain, Louvain-la-Neuve, Belgium (maxime.lagrange@uclouvain.be)
- 2National Institute for Nuclear Physics, University of Padova, Padova, Italy (tommaso.dorigo@gmail.com)
- 3National Institute for Nuclear Physics, University of Padova, Padova, Italy (giles.strong@outlook.com)
- 4Research Institute in Mathematics and Physics, Université Catholique de Louvain, Louvain-la-Neuve, Belgium (Andrea.Giammanco@cern.ch)
- 5Research Institute in Mathematics and Physics, Université Catholique de Louvain, Louvain-la-Neuve, Belgium (pietro.vischia@uclouvain.be)
- 6National Institute for Nuclear Physics, University of Padova, Padova, Italy (federico.nardi@fys.uio.no)
- 7National Institute for Nuclear Physics, University of Padova, Padova, Italy (federica.fanzago@pd.infn.it)
- 8Technical University of Munich, Munich, Germany (max.lamparth@tum.de)
We propose to employ differentiable programming techniques in order to construct a modular pipeline that models all the aspects of a muon tomography task, from the generation and interaction of cosmic ray muons with a parameterized detector and passive material, to the inference on the atomic number of the passive volume.
This enables the optimization of the detector parameters via gradient descent, to suggest optimal detector configurations, geometries, and specifications, subject to external constraints such as cost, detector size, and exposure time.
The eventual aim is to release the package open-source, to be used to guide the design of futur detectors for muon scattering and absorption imaging.
How to cite: Lagrange, M., Dorigo, T., Strong, G., Giammanco, A., Vischia, P., Fanzago, F., Nardi, F., and Lamparth, M.: TomOpt: Differentiable Muon-Tomography Optimization, EGU General Assembly 2022, Vienna, Austria, 23–27 May 2022, EGU22-9470, https://doi.org/10.5194/egusphere-egu22-9470, 2022.
