Applied Sciences,
Journal Year:
2024,
Volume and Issue:
14(18), P. 8520 - 8520
Published: Sept. 21, 2024
This
study
examines
the
impact
of
sensor
placement
and
multimodal
fusion
on
performance
a
Long
Short-Term
Memory
(LSTM)-based
model
for
human
activity
classification
taking
place
in
an
agricultural
harvesting
scenario
involving
human-robot
collaboration.
Data
were
collected
from
twenty
participants
performing
six
distinct
activities
using
five
wearable
inertial
measurement
units
placed
at
various
anatomical
locations.
The
signals
sensors
first
processed
to
eliminate
noise
then
input
into
LSTM
neural
network
recognizing
features
sequential
time-dependent
data.
Results
indicated
that
chest-mounted
provided
highest
F1-score
0.939,
representing
superior
over
other
placements
combinations
them.
Moreover,
magnetometer
surpassed
accelerometer
gyroscope,
highlighting
its
ability
capture
crucial
orientation
motion
data
related
investigated
activities.
However,
accelerometer,
showed
benefit
integrating
different
types
improve
accuracy.
emphasizes
effectiveness
strategic
optimizing
recognition,
thus
minimizing
requirements
computational
expenses,
resulting
cost-optimal
system
configuration.
Overall,
this
research
contributes
development
more
intelligent,
safe,
cost-effective
adaptive
synergistic
systems
can
be
integrated
variety
applications.
Journal of Construction Engineering and Management,
Journal Year:
2024,
Volume and Issue:
150(7)
Published: April 24, 2024
Worker
safety
and
productivity
the
factors
that
affect
them,
such
as
ergonomics,
are
essential
aspects
of
construction
projects.
The
application
ergonomics
identification
connections
between
workers
assigned
tasks
have
led
to
a
decrease
in
worker
injuries
discomfort,
beneficial
effects
on
productivity,
reduction
project
costs.
Nevertheless,
area
often
subjected
awkward
body
postures
repetitive
motions
cause
musculoskeletal
disorders,
turn
leading
delays
production.
As
systematic
widely
used
procedure
generates
final
document
or
form,
physical
demand
analysis
(PDA)
assesses
health
engaged
manufacturing
activities
proactively
evaluates
ergonomic
risks.
However,
gather
necessary
information,
traditional
PDA
methods
require
ergonomists
spend
significant
time
observing
interviewing
workers.
To
increase
speed
accuracy
PDA,
this
study
focuses
developing
framework
automatically
fill
posture-based
form
address
physiological
task
demands.
In
contrast
observation-based
approach,
proposed
uses
motion
capture
(MOCAP)
system
rule-based
expert
obtain
joint
angles
segment
positions
different
work
situations,
convert
measurements
objective
their
frequencies,
then
populate
forms.
is
tested
validated
both
laboratory
on-site
environments
by
comparing
generated
forms
with
filled
out
ergonomists.
results
indicate
MOCAP-/AI-based
automated
successfully
improves
performance
terms
accuracy,
consistency,
consumption.
Ultimately,
can
aid
design
job
goal
promoting
health,
safety,
workplace.
Applied System Innovation,
Journal Year:
2024,
Volume and Issue:
7(6), P. 103 - 103
Published: Oct. 28, 2024
The
comfort
of
a
worker
while
performing
any
activity
is
extremely
important.
If
that
extends
beyond
person’s
capacity
to
withstand
physical
and
psychological
stress,
the
may
suffer
from
both
mental
ailments.
Over
time,
if
stress
persists,
these
conditions
can
become
chronic
diseases
even
be
cause
workplace
accidents.
In
this
research,
methodology
was
developed
for
rapid
assessment
ergonomic
risks
calculating
level
in
workplace.
This
uses
artificial
intelligence
through
specific
algorithm
takes
into
account
number
factors
that,
when
combined,
have
significant
impact
on
workers.
To
achieve
more
accurate
simulation
work
situation
or
evaluate
an
ongoing
situation,
significantly
correlate
parameters,
we
used
logarithmic
calculation
formulas.
streamline
process,
software
performs
calculations,
conducts
risks,
estimates
level,
proposes
possible
measures
mitigate
effects
assist
diagnosing
neural
network
with
five
neurons
input
layer,
one
hidden
two
output
layer.
As
result,
most
situations,
industrial
field,
quickly
analyzed
evaluated
using
methodology.
use
new
analysis
diagnosis
tool,
implemented
research
technology,
beneficial
employers
Moreover,
further
developments
methodology,
achieved
by
increasing
relevant
parameters
ergonomics
integrating
advanced
systems,
aim
provide
high
precision
assessing
risk
comfort.
Applied Sciences,
Journal Year:
2024,
Volume and Issue:
14(18), P. 8520 - 8520
Published: Sept. 21, 2024
This
study
examines
the
impact
of
sensor
placement
and
multimodal
fusion
on
performance
a
Long
Short-Term
Memory
(LSTM)-based
model
for
human
activity
classification
taking
place
in
an
agricultural
harvesting
scenario
involving
human-robot
collaboration.
Data
were
collected
from
twenty
participants
performing
six
distinct
activities
using
five
wearable
inertial
measurement
units
placed
at
various
anatomical
locations.
The
signals
sensors
first
processed
to
eliminate
noise
then
input
into
LSTM
neural
network
recognizing
features
sequential
time-dependent
data.
Results
indicated
that
chest-mounted
provided
highest
F1-score
0.939,
representing
superior
over
other
placements
combinations
them.
Moreover,
magnetometer
surpassed
accelerometer
gyroscope,
highlighting
its
ability
capture
crucial
orientation
motion
data
related
investigated
activities.
However,
accelerometer,
showed
benefit
integrating
different
types
improve
accuracy.
emphasizes
effectiveness
strategic
optimizing
recognition,
thus
minimizing
requirements
computational
expenses,
resulting
cost-optimal
system
configuration.
Overall,
this
research
contributes
development
more
intelligent,
safe,
cost-effective
adaptive
synergistic
systems
can
be
integrated
variety
applications.
