EAA: Automating materials characterization with vision language model agents

TL;DR

EAA is a versatile system that automates complex microscopy workflows using vision-language models, integrating multimodal reasoning, tool use, and memory to improve efficiency and accessibility in experimental settings.

Contribution

The paper introduces EAA, a flexible agent system that automates microscopy workflows with multimodal reasoning, tool integration, and user interaction, advancing automation in experimental microscopy.

Findings

Automated zone plate focusing at a beamline.

Natural language feature search capability.

Reduced operational burden and increased efficiency.

Abstract

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.