Answers are grounded in this paper and cite their sources.
AI-enabled automated experimentation combines experimental platforms, analysis tools, and robots to support high-throughput screening and adaptive experimental decisions. These capabilities often rely on independently developed software, creating a need to coordinate execution while preserving the scientific context of results. We present MIRAL Cloud, an integration and orchestration system developed for the Material Innovation through Robotics and AI Laboratory (MIRAL) at Pacific Northwest National Laboratory. MIRAL Cloud exposes specialized laboratory capabilities as reusable operations within a shared workflow environment while retaining local instrument and procedure control. Three representative use cases illustrate complementary aspects of this approach: training and workflow development on a small-scale experimental platform, adaptive dissolution in which AI-based image analysis guides conditional solvent addition, and perceptionguided robotic sample handling. Together, these cases illustrate how experimental measurements, analytical responses, and placement outcomes remain linked to their originating requests and inform subsequent workflow steps across application-controlled experiments, commercial automation, and robotic operations. A browser workspace supports workflow configuration, execution monitoring, and inspection of results alongside supporting observations and source records. The contribution is a shared integration approach for coordinating heterogeneous capabilities within MIRAL, with reusable operation descriptions, connectors, and linked scientific records providing a basis for adaptation to other automation laboratories.
No takes yet. Share an insight, caveat, or question.
Zheng et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: