Mathematics,
Год журнала:
2024,
Номер
12(2), С. 339 - 339
Опубликована: Янв. 19, 2024
This
study
focuses
on
the
scheduling
problem
of
heterogeneous
unmanned
surface
vehicles
(USVs)
with
obstacle
avoidance
pretreatment.
The
goal
is
to
minimize
overall
maximum
completion
time
USVs.
First,
we
develop
a
mathematical
model
for
problem.
Second,
obstacles,
an
A*
algorithm
employed
generate
path
between
two
points
where
tasks
need
be
performed.
Third,
three
meta-heuristics,
i.e.,
simulated
annealing
(SA),
genetic
(GA),
and
harmony
search
(HS),
are
improved
solve
problems.
Based
problem-specific
knowledge,
nine
local
operators
designed
improve
performance
proposed
algorithms.
In
each
iteration,
Q-learning
strategies
used
select
high-quality
operators.
We
aim
meta-heuristics
by
using
Q-learning-based
Finally,
13
instances
different
scales
adopted
validate
effectiveness
strategies.
compare
classical
existing
meta-heuristics.
better
than
compared
ones.
results
comparisons
show
that
HS
second
Q-learning,
+
QL2,
exhibits
strongest
competitiveness
(the
smallest
mean
rank
value
1.00)
among
15
IEEE Transactions on Intelligent Transportation Systems,
Год журнала:
2024,
Номер
25(10), С. 15053 - 15064
Опубликована: Май 17, 2024
In
complex
and
variable
traffic
environments,
efficient
multi-objective
urban
light
scheduling
is
imperative.
However,
the
carbon
emission
problem
accompanying
delays
often
neglected
in
most
existing
literature.
This
study
focuses
on
problems
(MOUTLSP),
concerning
emissions
simultaneously.
First,
a
mathematical
model
firstly
developed
to
describe
MOUTLSP
minimize
vehicle
delays,
pedestrian
emissions.
Second,
three
well-known
meta-heuristics,
namely
genetic
algorithm
(GA),
particle
swarm
optimization
(PSO),
differential
evolution
(DE),
are
improved
solve
MOUTLSP.
Six
problem-feature-based
local
search
operators
(LSO)
designed
based
solution
structure
incorporated
into
iterative
process
of
meta-heuristics.
Third,
nature
utilized
design
two
novel
Q-learning-based
strategies
for
LSO
selection,
respectively.
The
selection
(QAS)
strategy
guides
non-dominated
solutions
obtain
good
trade-off
among
objectives
generates
high-quality
by
selecting
suitable
algorithms.
(QLSS)
employed
seek
premium
neighborhood
throughout
improving
convergence
speed.
effectiveness
improvement
verified
solving
11
instances
with
different
scales.
proposed
algorithms
compared
classical
some
state-of-the-art
problems.
experimental
results
comparisons
demonstrate
that
GA
$+$
QLSS,
variant
GA,
competitive
one.
research
proposes
new
ideas
Q-learning
assisted
evolutionary
firstly.
It
provides
strong
support
achieving
more
environmentally
friendly
management.
Wiley Interdisciplinary Reviews Data Mining and Knowledge Discovery,
Год журнала:
2024,
Номер
14(6)
Опубликована: Авг. 18, 2024
Abstract
This
paper
reviews
the
integration
of
Q‐learning
with
meta‐heuristic
algorithms
(QLMA)
over
last
20
years,
highlighting
its
success
in
solving
complex
optimization
problems.
We
focus
on
key
aspects
QLMA,
including
parameter
adaptation,
operator
selection,
and
balancing
global
exploration
local
exploitation.
QLMA
has
become
a
leading
solution
industries
like
energy,
power
systems,
engineering,
addressing
range
mathematical
challenges.
Looking
forward,
we
suggest
further
integration,
transfer
learning
strategies,
techniques
to
reduce
state
space.
article
is
categorized
under:
Technologies
>
Computational
Intelligence
Artificial
