Scalable Multi-Agent Lab Framework for Lab Optimization
A. Gilad Kusne, Austin McDannald
Autonomous materials research systems allow scientists to fail smarter, learn faster, and spend less resources in their studies. As these systems grow in number, capability, and complexity, a new challenge arises – how will they work together across large facilities? We explore one solution – a multi-agent laboratory-control framework. The frame-work is demonstrated with autonomous material science labs in mind, where information from diverse research campaigns can be combined to address scientific ques-tions. The framework can: 1) account for realistic re-source limits, e.g., equipment use, 2) allow for research-campaign-running machine-learning agents with diverse learning capabilities and goals, 3) facilitate multi-agent collaborations and teams. The MULTI-agent auTonomous fAcilities Scalable frameworK (MULTITASK) makes possible facility-wide simulations, including agent-instrument and agent-agent interactions. Through modu-larity, real-world facilities can come on-line in phases, with simulated instruments gradually replaced by real-world instruments. We hope MULTITASK opens new ar-eas of study in large-scale-autonomous and semi-autonomous research campaigns and facilities.
and McDannald, A.
Scalable Multi-Agent Lab Framework for Lab Optimization, Matter, [online], https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=935256
(Accessed November 28, 2023)