Inertial Navigation Meets Deep Learning: A Survey of Current Trends and Future Directions
Results in Engineering,
Год журнала:
2024,
Номер
unknown, С. 103565 - 103565
Опубликована: Дек. 1, 2024
Язык: Английский
A framework to assess the operational state of autonomous ships with multi-component degrading systems
Ocean Engineering,
Год журнала:
2025,
Номер
327, С. 121000 - 121000
Опубликована: Март 23, 2025
Язык: Английский
Ship Noise Characteristics in the Java Sea: A Preliminary Study on Underwater Noise Pollution in Indonesia
Research Square (Research Square),
Год журнала:
2025,
Номер
unknown
Опубликована: Май 16, 2025
Abstract
Indonesia
is
the
largest
archipelagic
nation
in
world,
facing
high
environmental
challenges
due
to
underwater
noise
generated
by
activities
from
various
types
of
ships.
Therefore,
this
study
aimed
examine
characteristics
(specifically
sound
pressure
level
(SPL)
and
frequency)
different
ships
operating
Java
Sea,
categorized
tonnage,
namely
under
30
GT,
30–100
exceeding
100
GT.
Using
a
calibrated
omnidirectional
hydrophone
system
alongside
synchronized
video
documentation,
acoustic
data
were
collected
analyzed
assess
intensity,
frequency,
duration.
The
results
showed
that
small
produced
higher
frequency
broadband
noise,
with
SPL
ranging
122
144
dB
re
1
µPa
based
on
speed.
Medium-sized
display
dominant
frequencies
kHz,
SPLs
related
engine
power
Large
ships,
such
as
ferries,
tugboats,
patrol
boats,
show
unique
spectral
profiles
influenced
type,
achieving
approximately
155
µPa.
This
significant
variability
emissions
type
operational
behavior
suggesting
need
for
mitigation
strategies
marine
policies
safeguard
Indonesia's
delicate
ecosystems.
Язык: Английский
A novel NSSA-LSTM-based forecasting model for ship delays in the Beijing-Hangzhou Grand Canal
Journal of Ocean Engineering and Marine Energy,
Год журнала:
2024,
Номер
unknown
Опубликована: Дек. 21, 2024
Язык: Английский
Selective maintenance of the complex system considering maintenance time uncertainty for system components with multiple repairpersons
Quality and Reliability Engineering International,
Год журнала:
2024,
Номер
unknown
Опубликована: Окт. 2, 2024
Abstract
This
research
presents
an
innovative
selected
maintenance
model
for
complex
systems
that
considers
the
uncertainty
in
time
(MT)
system
components
with
multiple
repairpersons.
The
computational
of
uncertain
MT
is
established.
An
imperfect
introduced,
which
has
many
levels
not
only
considering
do
nothing,
minimal
repair,
and
replacement
but
also
intermediate
levels.
Furthermore,
assignment
algorithm
repairpersons
proposed
to
addresses
problem
how
assign
tasks
order
minimize
MT.
And
it
innovatively
integrated
into
particle
swarm
optimization
(PSO)
solve
selective
model,
enables
heuristic
efficiently
effectiveness
advantages
are
verified
by
numerical
experiments.
Язык: Английский
Underwater Gyros Denoising Net (UGDN): A Learning-Based Gyros Denoising Method for Underwater Navigation
Journal of Marine Science and Engineering,
Год журнала:
2024,
Номер
12(10), С. 1874 - 1874
Опубликована: Окт. 18, 2024
Autonomous
Underwater
Vehicles
(AUVs)
are
widely
used
for
hydrological
monitoring,
underwater
exploration,
and
geological
surveys.
However,
AUVs
face
limitations
in
navigation
due
to
the
high
costs
associated
with
Strapdown
Inertial
Navigation
System
(SINS)
Doppler
Velocity
Log
(DVL),
hindering
development
of
low-cost
vehicles.
Micro
Electro
Mechanical
Measurement
Units
(MEMS
IMUs)
industry
their
low
cost
can
output
acceleration
angular
velocity,
making
them
suitable
as
an
Attitude
Heading
Reference
(AHRS)
poorly
calibrated
MEMS
IMUs
provide
inaccurate
leading
rapid
drift
orientation.
In
environments
where
cannot
use
GPS
position
correction,
this
have
severe
consequences.
To
address
issue,
paper
proposes
Gyros
Denoising
Net
(UGDN),
a
method
based
on
dilated
convolutions
LSTM
that
learns
extracts
spatiotemporal
features
IMU
sequences
dynamically
compensate
gyroscope’s
velocity
measurements,
reducing
attitude
heading
errors.
experimental
section
paper,
we
deployed
dataset
collected
from
field
trials
achieved
significant
results.
The
results
show
accuracy
data
denoised
by
UGDN
approaches
fiber-optic
SINS,
when
integrated
DVL,
it
serve
solution.
Язык: Английский