The :term:`U.S. Department of Energy (DOE)<DOE>`’s :term:`Artificial Intelligence (AI)<AI>` for Science report :cite:`AI:2020` outlines the need for intelligent systems, instruments, and facilities to enable science breakthroughs with autonomous experiments, self-driving laboratories, smart manufacturing, and :term:`AI`-driven design, discovery and evaluation :cite:`hanchen23scientific`. The :term:`DOE`’s Computational Facilities Research Workshop report :cite:`CF:2020` identifies intelligent systems/facilities as a challenge with enabling automation and reducing human-in-the-loop needs as a cross-cutting theme. The :term:`DOE`’s report on an :term:`Integrated Research Infrastructure (IRI)<IRI>` Architecture Blueprint Activity :cite:`IRI:2023` describes the core elements of the needed dynamic integration of experiment, observation, theory, modeling, simulation, visualization, :term:`machine learning (ML)<ML>`, :term:`AI`, and analysis.
Autonomous experiments, self-driving laboratories and smart manufacturing employ machine-in-the-loop intelligence for decision-making. Human-in-the-loop needs are reduced by an autonomous online control that collects experiment data, analyzes it, and takes appropriate operational actions to steer an ongoing or plan a next experiment. It may be assisted by an :term:`AI` that is trained online and/or offline with archived data and/or with synthetic data created by a digital twin. Analysis and decision making may also rely on rule-based approaches, causal or physics-based models, and advanced statistical methods. Human interaction for experiment planning, observation, and steering is performed through appropriate human-machine interfaces.
For example, both the rate and output of traditional materials synthesis and discovery are currently too slow and too small to efficiently provide needed advances. An :term:`autonomous robotic chemistry laboratory (ACL)<ACL>` (:numref:`intersect:arch:pat:introduction:acl`) can operate 24/7 with high precision to greatly accelerate materials discovery and innovation. It relies on the design of a laboratory utilizing robotic and autonomous tools for the manipulation of laboratory equipment and characterization tools. A robotic platform with three major components is used: a mobile base, a robotic arm, and software/characterization tools including integration/feedback with :term:`AI`.
An autonomous robotic chemistry laboratory that operates 24/7 using analysis of experimental data for the design of experiments.
A federated hardware/software ecosystem (:numref:`intersect:arch:pat:introduction:ecosystem`) for connecting instruments with edge and center computing resources is needed that autonomously collects, transfers, stores, processes, curates, and archives scientific data in common formats. It must be able to communicate with scientific instruments and computing and data resources for orchestration and ontrol across administrative domains, and with humans for critical decisions and feedback. Standardized communication and programming interfaces are needed that leverage community and custom software for scientific instruments, automation, workflows, and data transfer. Pluggability is required to permit quickly adaptable and deployable solutions, reuse of partial solutions for different use cases, and the use of digital twins, such as a virtual instrument, robot or experiment. This federated ecosystem needs to follow an open architecture standard to enable adoption.
The :term:`INTERSECT` ecosystem vision connects instruments with edge and center computing resources.
:term:`Oak Ridge National Laboratory's (ORNL's)<ORNL>` :term:`INTERconnected Science ECosysTem (INTERSECT)<INTERSECT>` Initiative offers an open federated hardware/software architecture for the laboratory of the future with a novel :ref:`intersect:arch:concept` that combines :ref:`intersect:arch:pat`, a :ref:`intersect:arch:sos` and a :ref:`intersect:arch:ms` for connecting scientific instruments, robot-controlled laboratories and edge/center computing/data resources to enable autonomous experiments, self-driving laboratories, smart manufacturing, and :term:`AI`-driven design, discovery and evaluation. :ref:`intersect:arch:examples` offer insight for applying this novel approach to real-world solutions. The :term:`DOE`'s recent efforts in an :ref:`intersect:arch:iri` are addressed as well.
.. youtube:: MQImdRf5wfc :align: center :width: 600