PLoS ONE,
Год журнала:
2024,
Номер
19(10), С. e0311297 - e0311297
Опубликована: Окт. 29, 2024
Greenhouse
cultivation
promotes
an
efficient
and
environmentally
friendly
agricultural
production
model,
significantly
enhancing
resource
sustainability
advancing
sustainable
agriculture.
Traditional
greenhouse
pollination
methods
are
inefficient
labor-intensive,
limiting
the
economic
benefits
of
pear
cultivation.
To
improve
efficiency
achieve
fully
automated
mechanized
operations,
innovative
design
method
for
drones
has
been
developed.
First,
criteria
were
extracted
using
Grounded
Theory
(GT),
Analytic
Hierarchy
Process
(AHP)
was
employed
to
determine
weight
user
demand
evaluation
indicators.
Next,
Quality
Function
Deployment
(QFD)
translated
needs
into
technical
requirements,
resulting
in
final
ranking
element
weights.
The
drone
then
designed
based
on
these
weighted
rankings,
yielding
optimal
solution.
This
quantifies
functional
requirements
product,
effectively
identifying
key
proposing
targeted
solutions.
Additionally,
it
provides
a
reference
other
highly
machinery
products.
Drones,
Год журнала:
2024,
Номер
8(11), С. 686 - 686
Опубликована: Ноя. 19, 2024
In
the
face
of
growing
challenges
in
modern
agriculture,
such
as
climate
change,
sustainable
resource
management,
and
food
security,
drones
are
emerging
essential
tools
for
transforming
precision
agriculture.
This
systematic
review,
based
on
an
in-depth
analysis
recent
scientific
literature
(2020–2024),
provides
a
comprehensive
synthesis
current
drone
applications
agricultural
sector,
primarily
focusing
studies
from
this
period
while
including
few
notable
exceptions
particular
interest.
Our
study
examines
detail
technological
advancements
systems,
innovative
aerial
platforms,
cutting-edge
multispectral
hyperspectral
sensors,
advanced
navigation
communication
systems.
We
analyze
diagnostic
applications,
crop
monitoring
mapping,
well
interventional
like
spraying
drone-assisted
seeding.
The
integration
artificial
intelligence
IoTs
analyzing
drone-collected
data
is
highlighted,
demonstrating
significant
improvements
early
disease
detection,
yield
estimation,
irrigation
management.
Specific
case
illustrate
effectiveness
various
crops,
viticulture
to
cereal
cultivation.
Despite
these
advancements,
we
identify
several
obstacles
widespread
adoption,
regulatory,
technological,
socio-economic
challenges.
particularly
emphasizes
need
harmonize
regulations
beyond
visual
line
sight
(BVLOS)
flights
improve
economic
accessibility
small-scale
farmers.
review
also
identifies
key
opportunities
future
research,
use
swarms,
improved
energy
autonomy,
development
more
sophisticated
decision-support
systems
integrating
data.
conclusion,
underscore
transformative
potential
technology
sustainable,
productive,
resilient
agriculture
global
21st
century,
highlighting
integrated
approach
combining
innovation,
adapted
policies,
farmer
training.
Frontiers in Robotics and AI,
Год журнала:
2024,
Номер
11
Опубликована: Окт. 17, 2024
Imitation
learning
(IL),
a
burgeoning
frontier
in
machine
learning,
holds
immense
promise
across
diverse
domains.
In
recent
years,
its
integration
into
robotics
has
sparked
significant
interest,
offering
substantial
advancements
autonomous
control
processes.
This
paper
presents
an
exhaustive
insight
focusing
on
the
implementation
of
imitation
techniques
agricultural
robotics.
The
survey
rigorously
examines
varied
research
endeavors
utilizing
to
address
pivotal
challenges.
Methodologically,
this
comprehensively
investigates
multifaceted
aspects
applications
encompasses
identification
tasks
that
can
potentially
be
addressed
through
detailed
analysis
specific
models
and
frameworks,
thorough
assessment
performance
metrics
employed
surveyed
studies.
Additionally,
it
includes
comparative
between
conventional
methodologies
realm
findings
derived
from
unveil
profound
insights
These
methods
are
highlighted
for
their
potential
significantly
improve
task
execution
dynamic
high-dimensional
action
spaces
prevalent
settings,
such
as
precision
farming.
Despite
promising
advancements,
discusses
considerable
challenges
data
quality,
environmental
variability,
computational
constraints
IL
must
overcome.
also
addresses
ethical
social
implications
implementing
technologies,
emphasizing
need
robust
policy
frameworks
manage
societal
impacts
automation.
hold
implications,
showcasing
revolutionize
processes
contributes
envisioning
innovative
tools
within
domain,
heightened
productivity
efficiency
robotic
systems.
It
underscores
remarkable
enhancements
various
processes,
signaling
transformative
trajectory
sector,
particularly
Agronomy,
Год журнала:
2023,
Номер
13(7), С. 1705 - 1705
Опубликована: Июнь 26, 2023
During
the
operation
of
agricultural
unmanned
aerial
vehicles
(UAVs)
in
orchards,
presence
power
poles
and
wires
pose
a
serious
threat
to
flight
safety,
can
even
lead
crashes.
Due
difficulty
directly
detecting
wires,
this
research
aimed
quickly
accurately
detect
wire
poles,
proposed
an
improved
Yolov5s
deep
learning
object
detection
algorithm
named
Yolov5s-Pole.
The
enhances
model’s
generalization
ability
robustness
by
applying
Mixup
data
augmentation
technique,
replaces
C3
module
with
GhostBottleneck
reduce
parameters
computational
complexity,
incorporates
Shuffle
Attention
(SA)
improve
its
focus
on
small
targets.
results
show
that
when
Yolov5s-Pole
model
was
used
for
accuracy,
recall,
mAP@50
were
0.803,
0.831,
0.838
respectively,
which
increased
0.5%,
10%,
9.2%
compared
original
model.
Additionally,
weights,
parameters,
GFLOPs
7.86
MB,
3,974,310,
9,
respectively.
Compared
model,
these
represent
compression
rates
42.2%,
43.4%,
43.3%,
time
single
image
using
4.2
ms,
good
under
different
lighting
conditions
(dark,
normal,
bright)
demonstrated.
is
suitable
deployment
UAVs’
onboard
equipment,
great
practical
significance
ensuring
efficiency
safety
UAVs.
Agriculture Ecosystems & Environment,
Год журнала:
2023,
Номер
362, С. 108815 - 108815
Опубликована: Дек. 6, 2023
Strawberry
(Fragaria
x
ananasa
Duch.)
is
the
most
economically
important
soft
fruit
worldwide.
While
self-
and
wind-pollination
possible
for
strawberry,
without
biotic
pollination
(animal
pollinators,
including
artificial
by
humans)
rate
of
strawberry
flowers
rarely
exceeds
60%
thus
production
decreased.
At
a
time
widely
recognized
decline
pollinators
increasing
global
demand
balanced
food,
we
need
comprehensive
understanding
worldwide
valuation
these
ecosystem
services.
In
this
paper,
use
transparent
systematic
review
process
to
detect
gaps
in
available
literature.
By
applying
multilevel
meta-analytic
models,
quantified
contribution
different
pollinator
types
fruits.
Our
showed
that
research
on
clearly
concentrated
European
countries,
with
limited
information
Asia
–
largest
producer.
Additionally,
detected
disproportions
across
topics,
stressing
further
preferences
between
cultivars
occurrence
managed
wild
bee
communities.
The
overall
estimate
benefit
crops
average
25%
reduction
weight
plants
not
receiving
treatment.
This
similar
regardless
pollinating
species.
mean
amount
globally
(2011–2020)
receivable
producers
from
purchasers
due
$5.36
billion
per
year.
Moreover,
demonstrated
are
dependent
or
set
43%
fewer
seeds
(fertilized
achenes)
than
biotically
pollinated
plants.
Such
huge
seed
can
cause
deformation
reduce
commercial
value
strawberries.
Together,
findings
indicate
critical
role
quality.
To
ensure
services
farms
produce
best-quality
fruits,
growers
should
take
action
sustain
healthy
communities
utilize
them
optimal
cultivar.
Abstract
Field-based
research
in
the
biological
sciences
encounters
several
challenges,
including
cost,
accessibility,
safety,
and
spatial
coverage.
Drones
have
emerged
as
a
transformative
technology
to
address
these
challenges
while
providing
less
intrusive
alternative
field
surveys.
Although
drones
mainly
been
used
for
high-resolution
image
collection,
their
capabilities
extend
beyond
mapping
production.
They
can
be
tailored
track
wildlife,
measure
environmental
parameters,
collect
physical
samples,
versatility
enables
researchers
tackle
variety
of
biodiversity
conservation
challenges.
In
this
article,
we
advocate
integrated
more
comprehensively
into
field-based
research,
from
site
reconnaissance
sampling,
interventions,
monitoring.
We
discuss
future
innovations
needed
harness
full
potential,
customized
instrumentation,
fit-for-purpose
software
apps,
better
integration
with
existing
online
databases.
also
support
leveraging
community
scientists
empowering
citizens
contribute
scientific
endeavors
promoting
stewardship
via
drones.
