- Physical Review Journals
All JournalsPhysics Magazine

Physical Review Research

Reuse & Permissions

It is not necessary to obtain permission to reuse this article or its components as it is available under the terms of the Creative Commons Attribution 4.0 International license. This license permits unrestricted use, distribution, and reproduction in any medium, provided attribution to the author(s) and the published article's title, journal citation, and DOI are maintained. Please note that some figures may have been included with permission from other third parties. It is your responsibility to obtain the proper permission from the rights holder directly for these figures.

Export citation

Export citation

Choose format for download:

Download Citation
  • Letter
  • Open Access

Agentic artificial intelligence for multistage physics experiments at a large-scale user facility particle accelerator

Thorsten Hellert*, Drew Bertwistle, Simon C. Leemann, Antonin Sulc, and Marco Venturini

  • *Contact author: thellert@lbl.gov
PDF

Phys. Rev. Research 8, L012017 – Published 16 January, 2026

DOI: https://doi.org/10.1103/jtqy-9jz1

Abstract

We present a language-model-driven agentic artificial intelligence (AI) system to autonomously execute multistage physics experiments on a production synchrotron light source. Implemented at the Advanced Light Source particle accelerator, the system translates natural language user prompts into structured execution plans that combine archive data retrieval, control-system channel resolution, automated script generation, controlled machine interaction, and analysis. In a representative machine physics task, we show that preparation time was reduced by 2 orders of magnitude relative to manual scripting even for a system expert, while operator-standard safety constraints were strictly upheld. Core architectural features, plan-first orchestration, bounded tool access, and dynamic capability selection, enable transparent, auditable execution with fully reproducible artifacts. These results establish a blueprint for the safe integration of agentic AI into accelerator experiments and demanding machine physics studies, as well as routine operations, with direct portability across accelerators worldwide and, more broadly, to other large-scale scientific infrastructures.

View figure in article

Physics Subject Headings (PhySH)

Article Text

References (42)

  1. T. Hellert, B. Flugstad, C. Sun, C. Steier, E. Wallén, F. Sannibale, G. Portmann, H. Nishimura, J. Weber, M. Venturini, M. Dach, S. C. Leemann, S. Omolayo, S. Borra, T. Scarvie, and T. Ford, in Proceedings of the 15th International Particle Accelerator Conference (JACoW Publishing, Geneva, Switzerland, 2024), paper TUPG37.
  2. C. Chen, et al., Strong electron-phonon coupling in magic-angle twisted bilayer graphene, Nature (London) 636, 342 (2024).
  3. S. Tan et al. Spontaneous formation of robust two-dimensional perovskite phases, Science 388, 639 (2025).
  4. S. K. Chandy, M. Lopez Luna, N. Z. Rustad, I. N. Zakaria, A. Siebert, S. Devlin, W. Li, M. Blum, and T. Head-Gordon, Ammonia synthesis under ambient conditions: Insights into water–nitrogen–magnetite interfaces, J. Am. Chem. Soc. 147, 24538 (2025).
  5. C. Y. Ralston, et al., ALS-ENABLE: Creating synergy and opportunity at the Advanced Light Source synchrotron structural biology beamlines, J. Synchrotron. Rad. 32, 1059 (2025).
  6. H. Wiedemann, Particle Accelerator Physics, Graduate Texts in Physics (Springer International Publishing, Cham, 2015).
  7. H. Damerau, Radio-frequency (RF) systems, Technical Report, CAS Course: Introduction to Accelerator Physics (CERN Accelerator School).
  8. J. Tanabe, Iron Dominated Electromagnets: Design, Fabrication, Assembly and Measurements (World Scientific, Singapore, 2005), p. 354.
  9. O. B. Malyshev, Vacuum in Particle Accelerators: Modelling, Design and Operation of Beam Vacuum Systems (Wiley-VCH Verlag GmbH & Co, Weinheim, Germany, 2019).
  10. M. G. Minty and F. Zimmermann, Measurement and Control of Charged Particle Beams, Particle Acceleration and Detection (Springer, Berlin, 2003).
  11. J. J. DiStefano III, A. R. Stubberud, and I. J. Williams, Schaums Outline of Feedback and Control Systems, 2nd ed. (McGraw‐Hill Professional, New York, 1997).
  12. S. J. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, 4th ed. (Prentice Hall, Hoboken, NJ, 2020).
  13. J. Wei, X. Wang, D. Schuurmans, M. Bosma, B. Ichter, F. Xia, E. Chi, Q. Le, and D. Zhou, in Advances in Neural Information Processing Systems (Curran Associates, Inc., 2022), Vol. 35, pp. 24824–24837.
  14. T. Schick, J. Dwivedi-Yu, R. Dessì, R. Raileanu, M. Lomeli, L. Zettlemoyer, N. Cancedda, and T. Scialom, Toolformer: Language models can teach themselves to use tools, Adv. Neural Inf. Process. Syst. 36, 68539 (2023).
  15. S. Yao, J. Zhao, D. Yu, N. Du, I. Shafran, K. Narasimhan, and Y. Cao, ReAct: Synergizing reasoning and acting in language models, arXiv:2210.03629.
  16. Q. Wu, G. Bansal, J. Zhang, Y. Wu, B. Li, E. Zhu, L. Jiang, X. Zhang, S. Zhang, J. Liu, A. H. Awadallah, R. W. White, D. Burger, and C. Wang, AutoGen: Enabling next-gen LLM applications via multi-agent conversation, arXiv:2308.08155.
  17. C. Packer, S. Wooders, K. Lin, V. Fang, S. G. Patil, and J. E. Gonzalez, MemGPT: Towards LLMs as operating systems, arXiv:2310.08560.
  18. LangGraph developers, LangGraph: A low-level orchestration framework for building resilient, stateful agents (2025), https://github.com/langchain-ai/langgraph.
  19. A. M. Bran, S. Cox, O. Schilter, C. Baldassari, A. D. White, and P. Schwaller, Augmenting large-language models with chemistry tools, Nat. Mach. Intell. 6, 525 (2024).
  20. D. A. Boiko, R. MacKnight, and G. Gomes, Autonomous chemical research with large language models, Nature (London) 624, 570 (2023).
  21. Y. Qu, K. Huang, M. Yin, K. Zhan, D. Liu, D. Yin, H. C. Cousins, W. A. Johnson, X. Wang, M. Shah, R. B. Altman, D. Zhou, M. Wang, and L. Cong, CRISPR-GPT for agentic automation of gene-editing experiments, Nat. Biomed. Eng. (2025).
  22. S. Mathur, N. V. der Vleuten, K. G. Yager, and E. H. R. Tsai, VISION: A modular AI assistant for natural human-instrument interaction at scientific user facilities, Mach. Learn.: Sci. Technol. 6, 025051 (2025).
  23. F. Mayet, GAIA: A general AI assistant for intelligent accelerator operations, arXiv:2405.01359.
  24. J. Kaiser, A. Lauscher, and A. Eichler, Large language models for human-machine collaborative particle accelerator tuning through natural language, Sci. Adv. 11, eadr4173 (2025).
  25. A. Sulc, T. Hellert, R. Kammering, H. Houscher, and J. St. John, Towards agentic AI on particle accelerators, arXiv:2409.06336.
  26. L. R. Dalesio, J. O. Hill, M. Kraimer, S. Lewis, D. Murray, S. Hunt, W. Watson, M. Clausen, and J. Dalesio, The experimental physics and industrial control system architecture: Past, present, and future, Nucl. Instrum. Methods Phys. Res. Sect. A 352, 179 (1994).
  27. M. Shankar, M. Davidsaver, M. Konrad, and L. Li, The EPICS archiver appliance, in Proceedings of the 15th International Conference on Accelerator and Large Experimental Physics Control Systems (ICALEPCS 2015) (JACoW, Melbourne, Australia, 2015), pp. 761–764.
  28. T. Kluyver, B. Ragan-Kelley, F. Pérez, B. Granger, M. Bussonnier, J. Frederic, K. Kelley, J. Hamrick, J. Grout, S. Corlay, P. Ivanov, D. Avila, S. Abdalla, and C. Willing, in Positioning and Power in Academic Publishing: Players, Agents and Agendas, edited by F. Loizides and B. Schmidt (IOS Press, Amsterdam, Netherlands, 2016), pp. 87–90.
  29. T. Hellert, J. Montenegro, and A. Sulc, Osprey: A scalable framework for the orchestration of agentic systems, arXiv:2508.15066.
  30. ALS Assistant Developers, ALS assistant repo, (2025), https://github.com/als-apg/als_assistant.
  31. Open WebUI (2023), https://github.com/open-webui/open-webui.
  32. B. E. Granger and F. Pérez, Jupyter: Thinking and storytelling with code and data, Comput. Sci. Eng. 237 (2021).
  33. Ollama Team, Ollama: Get up and running with large language models (2023), https://github.com/ollama/ollama.
  34. Lawrence Berkeley National Laboratory, Science IT Group, CBORG AI portal, Lawrence Berkeley National Laboratory AI service platform (2024), https://cborg.lbl.gov/.
  35. OpenAI, ChatGPTt language model family (2022), https://openai.com/chatgpt, accessed 15 September 2025.
  36. Anthropic, Claude language model family (2023), https://docs.anthropic.com/en/docs/about-claude/models/overview.
  37. Google DeepMind, Gemini language model family (2023), https://deepmind.google/models/gemini/.
  38. Red Hat, Inc., Podman: A daemonless OCI-compliant container engine, version 4.x (2024), https://podman.io/.
  39. K. Shroff, T. Ashwarya, T. Ford, K.-U. Kasemir, R. Lange, and G. Weiss, Phoebus tools and services, in JACoW, Proc. ICALEPCS2023 (2023).
  40. Anthropic, Claude agent SDK for Python, GitHub, 2025, https://github.com/anthropics/claude-agent-sdk, documentation: https://docs.anthropic.com/en/docs/claude-code/sdk/sdk-python.
  41. Undulators, Wigglers and Their Applications, edited by H. Onuki and P. Elleaume (CRC Press, Boca Raton, 2003).
  42. G. Portmann, J. Corbett, and A. Terebilo, An accelerator control middle layer using MATLAB, Conf. Proc. C 0505161, 4009 (2005).

Outline

Information

Phys. Rev. Research 8, L012017– Published 16 January, 2026
  • Received 21 September 2025
  • Accepted 5 December 2025
Crossmark - Check for updates button
Creative Commons Logo

Published by the American Physical Society under the terms of the Creative Commons Attribution 4.0 International license. Further distribution of this work must maintain attribution to the author(s) and the published article's title, journal citation, and DOI.

Published by the American Physical Society

Sign In to Your Journals Account

Filter

Filter

Article Lookup

Enter a citation