Compliance
with
good
research
data
management
practices
means
trust
in
the
integrity
of
data,
and
it
is
achievable
by
a
full
control
gathering
process.
In
this
work,
we
demonstrate
tooling
which
bridges
these
two
aspects,
illustrate
its
use
case
study
automated
battery
cycling.
We
successfully
interface
off-the-shelf
cycling
hardware
computational
workflow
software
AiiDA,
allowing
us
to
experiments,
while
ensuring
tracking
provenance.
design
user
interfaces
compatible
tooling,
span
inventory,
experiment
design,
result
analysis
stages.
Other
features,
including
monitoring
workflows
import
externally
generated
legacy
are
also
implemented.
Finally,
stack
required
for
work
made
available
set
open-source
packages.
Digital Discovery,
Journal Year:
2023,
Volume and Issue:
3(1), P. 23 - 33
Published: Dec. 6, 2023
The
ASLLA
Symposium
focused
on
accelerating
chemical
science
with
AI.
Discussions
data,
new
applications,
algorithms,
and
education
were
summarized.
Recommendations
for
researchers,
educators,
academic
bodies
provided.
Science,
Journal Year:
2024,
Volume and Issue:
384(6697)
Published: May 16, 2024
Contemporary
materials
discovery
requires
intricate
sequences
of
synthesis,
formulation,
and
characterization
that
often
span
multiple
locations
with
specialized
expertise
or
instrumentation.
To
accelerate
these
workflows,
we
present
a
cloud-based
strategy
enabled
delocalized
asynchronous
design-make-test-analyze
cycles.
We
showcased
this
approach
through
the
exploration
molecular
gain
for
organic
solid-state
lasers
as
frontier
application
in
optoelectronics.
Distributed
robotic
synthesis
in-line
property
characterization,
orchestrated
by
artificial
intelligence
experiment
planner,
resulted
21
new
state-of-the-art
materials.
Gram-scale
ultimately
allowed
verification
best-in-class
stimulated
emission
thin-film
device.
Demonstrating
integration
five
laboratories
across
globe,
workflow
provides
blueprint
delocalizing-and
democratizing-scientific
discovery.
Nature Communications,
Journal Year:
2024,
Volume and Issue:
15(1)
Published: Jan. 23, 2024
Abstract
The
ability
to
integrate
resources
and
share
knowledge
across
organisations
empowers
scientists
expedite
the
scientific
discovery
process.
This
is
especially
crucial
in
addressing
emerging
global
challenges
that
require
solutions.
In
this
work,
we
develop
an
architecture
for
distributed
self-driving
laboratories
within
World
Avatar
project,
which
seeks
create
all-encompassing
digital
twin
based
on
a
dynamic
graph.
We
employ
ontologies
capture
data
material
flows
design-make-test-analyse
cycles,
utilising
autonomous
agents
as
executable
components
carry
out
experimentation
workflow.
Data
provenance
recorded
ensure
its
findability,
accessibility,
interoperability,
reusability.
demonstrate
practical
application
of
our
framework
by
linking
two
robots
Cambridge
Singapore
collaborative
closed-loop
optimisation
pharmaceutically-relevant
aldol
condensation
reaction
real-time.
graph
autonomously
evolves
toward
scientist’s
research
goals,
with
effectively
generating
Pareto
front
cost-yield
three
days.
Self-driving
laboratories
(SDLs)
are
next-generation
research
and
development
platforms
for
closed-loop,
autonomous
experimentation
that
combine
ideas
from
artificial
intelligence,
robotics,
high-performance
computing.
A
critical
component
of
SDLs
is
the
decision-making
algorithm
used
to
prioritize
experiments
be
performed.
This
SDL
“brain”
often
relies
on
optimization
strategies
guided
by
machine
learning
models,
such
as
Bayesian
optimization.
However,
diversity
hardware
constraints
scientific
questions
being
tackled
require
availability
a
set
flexible
algorithms
have
yet
implemented
in
single
software
tool.
Here,
we
report
Atlas,
an
application-agnostic
Python
library
specifically
tailored
needs
SDLs.
Atlas
provides
facile
access
state-of-the-art,
model-based
algorithms—including
mixed-parameter,
multi-objective,
constrained,
robust,
multi-fidelity,
meta-learning,
molecular
optimization—as
all-in-one
tool
expected
suit
majority
specialized
needs.
After
brief
description
its
core
capabilities,
demonstrate
Atlas’
utility
optimizing
oxidation
potential
metal
complexes
with
electrochemical
platform.
We
expect
expand
breadth
design
discovery
problems
natural
sciences
immediately
addressable
Digital Discovery,
Journal Year:
2023,
Volume and Issue:
2(6), P. 1806 - 1812
Published: Jan. 1, 2023
Human
researchers
multi-task,
collaborate,
and
share
resources.
HELAO-async
is
a
multi-workflow
automation
software
that
helps
realize
these
attributes
in
materials
acceleration
platforms.
Contemporary
materials
discovery
requires
intricate
sequences
of
synthesis,
formulation
and
characterization
that
often
span
multiple
locations
with
specialized
expertise
or
instrumentation.
To
accelerate
these
workflows,
we
present
a
cloud-based
strategy
enables
delocalized
asynchronous
design–make–test–analyze
cycles.
We
showcase
this
approach
through
the
exploration
molecular
gain
for
organic
solid-state
lasers
as
frontier
application
in
optoelectronics.
Distributed
robotic
synthesis
in-line
property
characterization,
orchestrated
by
AI
experiment
planner,
resulted
21
new
state-of-the-art
materials.
Automated
gram-scale
ultimately
allowed
verification
best-in-class
stimulated
emission
thin-film
device.
Demonstrating
integration
five
laboratories
across
globe,
workflow
provides
blueprint
delocalizing
–
democratizing
scientific
discovery.
Machine Learning Science and Technology,
Journal Year:
2024,
Volume and Issue:
5(2), P. 02LT01 - 02LT01
Published: May 31, 2024
Abstract
With
the
advent
of
large
language
models
(LLMs),
in
both
open
source
and
proprietary
domains,
attention
is
turning
to
how
exploit
such
artificial
intelligence
(AI)
systems
assisting
complex
scientific
tasks,
as
material
synthesis,
characterization,
analysis
discovery.
Here,
we
explore
utility
LLMs,
particularly
ChatGPT4,
combination
with
application
program
interfaces
(APIs)
tasks
experimental
design,
programming
workflows,
data
scanning
probe
microscopy,
using
in-house
developed
APIs
given
by
a
commercial
vendor
for
instrument
control.
We
find
that
LLM
can
be
especially
useful
converting
ideations
workflows
executable
code
on
microscope
APIs.
Beyond
generation,
GPT4
capable
analyzing
microscopy
images
generic
sense.
At
same
time,
suffers
from
an
inability
extend
beyond
basic
analyses
more
in-depth
technical
design.
argue
specifically
fine-tuned
individual
domains
potentially
better
interface
human
experts
workflows.
Such
synergy
between
expertise
efficiency
experimentation
new
doors
accelerating
research,
enabling
effective
protocols
sharing
community.
Digital Discovery,
Journal Year:
2024,
Volume and Issue:
3(7), P. 1273 - 1279
Published: Jan. 1, 2024
We
share
the
results
of
a
survey
on
automation
and
autonomy
in
materials
science
labs,
which
highlight
variety
researcher
challenges
motivations.
also
propose
framework
for
levels
laboratory
from
L0
to
L5.
Digital Discovery,
Journal Year:
2023,
Volume and Issue:
2(6), P. 1980 - 1998
Published: Jan. 1, 2023
Advances
in
robotic
automation,
high-performance
computing,
and
artificial
intelligence
encourage
us
to
propose
large,
general-purpose
science
factories
with
the
scale
needed
tackle
large
discovery
problems
support
thousands
of
scientists.
