Saved in:
Bibliographic Details
Main Authors: Du, Ming, Luo, Yanqi, Banerjee, Srutarshi, Wojcik, Michael, Popovic, Jelena, Cherukara, Mathew J.
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.15294
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912909480689664
author Du, Ming
Luo, Yanqi
Banerjee, Srutarshi
Wojcik, Michael
Popovic, Jelena
Cherukara, Mathew J.
author_facet Du, Ming
Luo, Yanqi
Banerjee, Srutarshi
Wojcik, Michael
Popovic, Jelena
Cherukara, Mathew J.
contents We present Experiment Automation Agents (EAA), a vision-language-model-driven agentic system designed to automate complex experimental microscopy workflows. EAA integrates multimodal reasoning, tool-augmented action, and optional long-term memory to support both autonomous procedures and interactive user-guided measurements. Built on a flexible task-manager architecture, the system enables workflows ranging from fully agent-driven automation to logic-defined routines that embed localized LLM queries. EAA further provides a modern tool ecosystem with two-way compatibility for Model Context Protocol (MCP), allowing instrument-control tools to be consumed or served across applications. We demonstrate EAA at an imaging beamline at the Advanced Photon Source, including automated zone plate focusing, natural language-described feature search, and interactive data acquisition. These results illustrate how vision-capable agents can enhance beamline efficiency, reduce operational burden, and lower the expertise barrier for users.
format Preprint
id arxiv_https___arxiv_org_abs_2602_15294
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle EAA: Automating materials characterization with vision language model agents
Du, Ming
Luo, Yanqi
Banerjee, Srutarshi
Wojcik, Michael
Popovic, Jelena
Cherukara, Mathew J.
Artificial Intelligence
68T42
I.2.1
We present Experiment Automation Agents (EAA), a vision-language-model-driven agentic system designed to automate complex experimental microscopy workflows. EAA integrates multimodal reasoning, tool-augmented action, and optional long-term memory to support both autonomous procedures and interactive user-guided measurements. Built on a flexible task-manager architecture, the system enables workflows ranging from fully agent-driven automation to logic-defined routines that embed localized LLM queries. EAA further provides a modern tool ecosystem with two-way compatibility for Model Context Protocol (MCP), allowing instrument-control tools to be consumed or served across applications. We demonstrate EAA at an imaging beamline at the Advanced Photon Source, including automated zone plate focusing, natural language-described feature search, and interactive data acquisition. These results illustrate how vision-capable agents can enhance beamline efficiency, reduce operational burden, and lower the expertise barrier for users.
title EAA: Automating materials characterization with vision language model agents
topic Artificial Intelligence
68T42
I.2.1
url https://arxiv.org/abs/2602.15294