Path planning optimization in unmanned aerial vehicles using meta-heuristic algorithms: a systematic review

Abstract

Unmanned aerial vehicles (UAVs) have recently been increasingly popular in various areas, fields, and applications. Military, disaster management, rescue operations, public services, agriculture, and various other areas are examples. As a result, UAV path planning is concerned with determining the optimal path from the source to the destination while avoiding collisions with lowering the cost of time, energy, and other resources. This review aims to assort academic studies on the path planning optimization in UAV using meta-heuristic algorithms, summarize the results of each optimization algorithm, and extend the understanding of the current state of the path planning in UAV in the meta-heuristic optimization field. For this purpose, we implemented a broad, automated search using Boolean and snowballing searching methods to find academic works on path planning in UAVs. Studies and papers have been distinguished, and the following information was obtained and aggregated from each article: authors, publication’s year, the journal name or the conference name, proposed algorithms, the aim of the study, the outcome, and the quality of each study. According to the findings, the meta-heuristic algorithm is a standard optimization method for tackling single and multi-objective problems. Besides, the findings show that meta-heuristic algorithms have a great compact on the path planning optimization in UAVs, and there is good progress in this field. However, the problem still exists mainly in complex and dynamic environments, on battlefields, in rescue missions, mobile obstacles, and with multiple UAVs.

Full text

Vol.: (0123456789)
1 3
Environ Monit Assess (2023) 195:30
https://doi.org/10.1007/s10661-022-10590-y
Path planning optimization inunmanned aerial vehicles
using meta‑heuristic algorithms: asystematic review
HazhaSaeedYahia· AminSalihMohammed
Received: 18 October 2021 / Accepted: 22 January 2022
© The Author(s), under exclusive licence to Springer Nature Switzerland AG 2022
the path planning in UAV in the meta-heuristic opti-
mization field. For this purpose, we implemented a
broad, automated search using Boolean and snow-
balling searching methods to find academic works on
path planning in UAVs. Studies and papers have been
distinguished, and the following information was
obtained and aggregated from each article: authors,
publication’s year, the journal name or the confer-
ence name, proposed algorithms, the aim of the study,
the outcome, and the quality of each study. Accord-
ing to the findings, the meta-heuristic algorithm is a
standard optimization method for tackling single and
multi-objective problems. Besides, the findings show
that meta-heuristic algorithms have a great compact
on the path planning optimization in UAVs, and there
is good progress in this field. However, the problem
still exists mainly in complex and dynamic environ-
ments, on battlefields, in rescue missions, mobile
obstacles, and with multiple UAVs.
Keywords Unmanned aerial vehicles· Path
planning· Meta-heuristic algorithms· Environment
monitoring
Introduction
Unmanned aerial vehicles (UAV)s, likewise known
as drones, unmanned aircraft, and pilotless aircraft,
can be defined as a powered flying machine or vehi-
cle without an onboard human pilot and operator with
Abstract Unmanned aerial vehicles (UAVs) have
recently been increasingly popular in various areas,
fields, and applications. Military, disaster manage-
ment, rescue operations, public services, agriculture,
and various other areas are examples. As a result,
UAV path planning is concerned with determining
the optimal path from the source to the destination
while avoiding collisions with lowering the cost of
time, energy, and other resources. This review aims
to assort academic studies on the path planning opti-
mization in UAV using meta-heuristic algorithms,
summarize the results of each optimization algorithm,
and extend the understanding of the current state of
Supplementary Information The online version
contains supplementary material available at https:// doi.
org/ 10. 1007/ s10661- 022- 10590-y.
H.S.Yahia(*)
Department ofInformation Technology, Lebanese French
University, Erbil, Iraq
e-mail: hazha.yahia@hotmail.com
H.S.Yahia
Department ofInformation Technology, Duhok
Polytechnic University, Duhok, Iraq
A.S.Mohammed
Department ofComputer Engineering, Lebanese French
University, Erbil, Iraq
A.S.Mohammed
Department ofSoftware andInformatics, Salahaddin
University, Erbil, Iraq
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 2 of 28
Vol:. (1234567890)
the ability to operate remotely or autonomously with
the use of a camera, a sensor, communication’s equip-
ment, or other payloads (Valavanis & Vachtsevanos,
2015). Initially, UAVs were employed as weapons
in the military field for remotely guided aerial mis-
siles (Blom, 2006; Khamseh etal., 2009). However, it
recently became extensively used in many fields and
applications, including military surveillance, recon-
naissance, and combat operations, to reduce human
casualties in civil fields such as forest fire monitor-
ing, rescue operations, agriculture, hazardous envi-
ronments, mapping, geology, tracking operations, and
many other areas (Jun & D’Andrea, 2003; Valavanis,
2017; Zheng etal., 2018).
As UAVs are used in different fields and applica-
tions, the operational missions are various, and each
requires a particular UAV with different specifica-
tions. Therefore, UAVs have different sizes ranging
from a macro or nano drones to small-tactical drones,
medium-sized reconnaissance, and large combat and
surveillance UAVs. Besides, they have a different
operational range as long-range, mid-range, and short-
range (Valavanis & Vachtsevanos, 2015; Marqués,
2016). UAVs are important since they do not require
humans or pilots because they are pilotless vehicles.
Many occupations, particularly those in the military,
warfare, and forest fire detection and rescue, are haz-
ardous. UAVs, on the other hand, are crucial because
they can operate in a wide range of environments, and
their applications are promising and cost-effective
(Danancier etal., 2019).
The UAV usually operates through essential com-
ponents, including the actual UAV (including an
airframe, navigation system, power system, and pay-
load), the ground controller, and the communication
platform connecting the two of them (Giordan etal.,
2020). The UAV functions can be controlled using
different ways such as onboard systems, human oper-
ator, and so on (Valavanis & Vachtsevanos, 2015).
Path planning for unmanned aerial vehicles
(UAVs) is an optimization problem, and it has played
an important part in recent research. Generally, path
planning is to search for the optimal or near-optimal
path from the source or initial point to the distention
or the target (Tsourdos et al., 2010). The optimal
path maybe is the shortest, avoiding obstacles and
collision-free in uncertain environments and meet-
ing its missions (Valavanis & Vachtsevanos, 2015;
Rudas & Tar, 2010). To find an optimal path, UAVs
must create a map (if the map is not existing), local-
ize its current position, and deal with its position if
the surrounding environment is uncertain and must
avoid obstacles. The literature classified the path
planning methods into different categories; first is
based on the type, which includes global path plan-
ning and local path planning. Second is based on
the time domain, which includes offline and online
time domains. Third is based on the space domain
and has two types: two-dimensional (2D) and three-
dimensional environments (3D). Fourth is based on
the optimization methods, such as map-based meth-
ods, potential-field methods, mathematical-based
methods, and evolutionary-based planning methods.
Finally, it is based on the algorithms, including con-
ventional algorithms, cell-based algorithms, model-
based algorithms, and learning-based algorithms
(Yang etal., 2016), as shown in the below figure.
Fred glover coined the term “meta-heuristic”
and combined the Greek prefix meta- (metá, means
higher-level) with heuristic (has Greek roots and
means heuriskein or euriskein). Meta-heuristics are
typical of a stochastic nature (Lukač, 2011) and can
be defined as a problem-independent, high-level
algorithmic framework that provides a set of recom-
mendations or strategies for creating heuristic opti-
mization algorithms (Sörensen, 1986). Meta-heuristic
algorithms can be divided into four categories: evolu-
tionary algorithms, physics-based algorithms, swarm-
based algorithms, and hybrid meta-heuristic algo-
rithms (Khalilpourazari et al., 2019). Evolutionary
algorithms are the most common type of meta-heuristic
algorithm.
This paper aims to provide a detailed analysis of the
path planning problems in UAVs using meta-heuristic
algorithms. Most of the reviewed papers optimized
the path planning based on different methods as map-
based methods, potential field methods, mathematical
planning methods, and evolutionary-based methods.
Collecting (302) research papers have been collected
from accessible databases with the expectation that the
collected paper will provide new insight into the path
planning optimization in UAVs. The paper’s organiza-
tion was presented as follows: the “Methodology” sec-
tion presents the methodology used for this systematic
review; the “Results and analysis” section displays the
discussions, results, and analysis; the “Discussion”
section shows the gaps in the literature related to this
subject, and the last section is the conclusion.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 3 of 28 30
Vol.: (0123456789)
Methodology
A systematic review employs a well-defined method-
ology to elicit researches from recent studies pertinent
to this research topic. This review is primarily based
on the Kitchenham method and Denyer and Tranfield
(Denyer & Tranfield, 2009; Kitchenham etal., 2009).
A specific protocol has been designed for this review;
the protocol consists of six phases: research question,
search strategy, inclusion/exclusion criteria, selection
criteria, data extraction, and data synthesis. In the
subsequent section, each stage has been discussed in
detail.
Research questions
In the first phase, based on the aim of this review, the
main question is specified. This review aims to clas-
sify the researches in UAV path planning optimiza-
tion and study the details and characteristics of each
proposed meta-heuristic algorithm. Next, it finds the
advantage results and limitations of each algorithm.
Accordingly, the main question specified as to “How
to optimize UAV path planning using meta-heuristic
algorithms?” The question is broad and general and
can be broken down into more sub-questions that
make understanding the main problem clearer and
finding the limitations and gaps in this area more
accessible. The following are the sub-questions of the
main question:
RQ1: What are the types of meta-heuristic algo-
rithm?
RQ2: In what applications are these algorithms
tested?
RQ3: Does the UAV operates in a 2D or 3D envi-
ronment?
RQ4: Does the UAV operates in offline or online
time domain?
RQ5: Does UAV use maps or sensors to reach the
destination?
RQ6: What are the type of obstacles have been
used in the environment and their numbers?
RQ7: Does the algorithm tested on a single UAV
or multiple UAVs?
Each of the questions above has been addressed in
the analysis section.
Search strategy
In the second phase and based on the research ques-
tion, searching for literature was accomplished in
a set of popular search engines and online avail-
able computer science databases. Including, IEEE
Xplore,1 Elsevier,2 SpringerLink,3 MDPI,4 ACM
Digital Library,5 Hindawi,6 and Google Scholar.7 We
used special search techniques, Boolean search and
snowball searching methods, to search for the related
researches in the searching engines. The Boolean
search approach converts the research question into
strings and keywords and retrieves researches from
searching databases. Based on these strings and key-
words, a list of synonyms and related words will be
created, connecting the list with the Boolean opera-
tors (AND, OR, and NOT) for searching and retriev-
ing researches (Aliyu, 2017). In this work, the AND
operation has been used for searching with Boolean
search. Next, snowballing search method has been
employed to find highly cited works in the field of
UAV path planning optimization (Wohlin, 2014),
beginning with backward snowballing, which entails
using the reference list of each retrieved article and
extracting or retrieving the research that satisfies this
review’s inclusion and exclusion criteria, while omit-
ting the papers studied previously. Hence, Google
Scholar has been used to find additional works that
cited the reviewed research during the forward snow-
balling. Then, each research that cites the paper
is scrutinized. Following the selection of citing
research, the article passes through inclusion and
exclusion for retrieval.
For each search engine, different search keywords
have been used due to the differences in the databases
and their capabilities. In the Boolean search process,
different phrases and keywords related to the subject
are used and categorized into three main categories; the
first category includes keywords “UAV path planning,”
“path planning in Robotics,” “path planning problem in
UAV,” and “path planning in Drones” in general. Then
1 https:// ieeex plore. ieee. org/
2 https:// www. elsev ier. com/
3 https:// www. sprin ger. com/
4 https:// www. mdpi. com/
5 https:// dl. acm. org/
6 https:// www. hinda wi. com/
7 https:// schol ar. google. com. tw/
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 4 of 28
Vol:. (1234567890)
the search keywords narrowed to “UAV path planning
optimization.” The second category includes keywords
as “Meta-heuristic algorithms for UAV path planning.”
Finally, the last category includes “meta-heuristic UAV
path planning” and “meta-heuristic optimization path
planning in UAV.” The second part of the search has
been done by using forward and backward snowballing.
For the forward, we depend on the citations of the papers
in Google Scholar and for the backward snowballing, we
depend on the reference lists of the researches. Based on
the two different search categories, the findings reached
340 researches and conference papers.
Inclusion/exclusion criteria
Although much research has been founded in the
searching process, not all are useful for this review.
Therefore, in the third phase, the collected researches
have been filtered based on the review question.
Besides, some restrictions have been defined in the
third phase. The following are the criteria of the
restrictions that we applied for filtering and shortlist-
ing the collected researches:
Inclusion criteria are as follows:
• All papers are entirely written in English.
• All papers must be published by January 2010 or
later.
• Papers must be regular papers and peer-reviewed
(journal researches, conference papers).
• There should be no redundant research researches
preserved. Each article must concentrate on origi-
nal research work.
• The papers’ content is identical to the subject of
the systematic review “Optimizing path planning
in UAV using meta-heuristic algorithms.”
Exclusion criteria are as follows:
• The researches are using other algorithms rather
than meta-heuristics for optimizing path planning.
• The researches are published in fake journals or
contain scientific bias.
• Papers from literary artifacts are not permitted,
such as chapters of books, editorials, or remarks.
In constraining the search results, these crite-
ria aided in the refinement of the search results and
limited the studies to the relevance to the objective of
this review only. Hence, 302 researches were filtered,
and only 70 relevant studies remain.
Selection criteria
The fourth phase involves manually evaluating the
shortlisted researches to demonstrate their relevance
to the aim and objective of this review. In the previ-
ous phase, the article exclusion was based on the
titles only, but here it is based on titles and abstracts
for the selected researches in the previous phase. The
abstracts of each article are meticulously examined
and analyzed to decide its relevance. If the abstracts
or written text does not contain the relevant key-
words, the article is rejected. Following the analysis
procedure, only papers that perfectly suit the review’s
scope are kept. This procedure resulted in a con-
densed list of 30 relevant research publications cited
in the current investigation; Table1 shows the num-
ber of retrieved article and then number of related
retrieved researches after implementing the inclusion
and exclusion criteria. The “other” search engines
includes Simulation: SAGE Journals, The Scientific
World Journal, Proceedings of the 33rd Chinese Con‑
trol Conference, 2015 International Conference on
Unmanned Aircraft Systems (ICUAS), International
Journal of Advanced Computer Science and Applica‑
tions (IJACSA), and International Journal of Inno‑
vative Computing, Information and Control (ICIC).
Besides, the flowchart diagram shows the process of
searching in databases for relevant research followed
by the inclusion and exclusion process and shows the
number of accepted or rejected researches, as shown
in Fig.1.
Data extraction
In the fourth phase, the information collection
addresses the review question and study criteria.
The data extraction for this review is defined as a
set of specific values extracted from each article.
These values include aim and objective of the study,
proposed algorithm, type and class of the proposed
algorithm, algorithm base (for example, the algo-
rithm is based on genetic algorithm (GA)), set of
parameters each study are based on, and summariz-
ing the results of each study. As the data extraction
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 5 of 28 30
Vol.: (0123456789)
must include answers to the review question, data
collection forms have been created to provide stand-
ard information, including the title and author/s of
the article, date of publishing with the name of the
journal, country, proposed algorithm and its classi-
fication, and many other fields. For more informa-
tion, see Appendix A, the data extraction form.
Data synthesis
In this phase, compiling and summarizing the results
of included studies have been done. Usually, the data
synthesis can be descriptive or quantitative, so we
based on a descriptive summary for the results in
this systematic review. The extracted information is
Table 1 Number of retrieved researches after implementing the inclusion and exclusion criteria
Search engines No. of retrieved
researches in search
engines
No. of retrieved researches after
implementing the inclusion and
exclusion criteria
No. of retrieved
researches based
on titles
No. of retrieved related
researches based on title and
abstract
IEEE Xplore 125 98 32 10
Elsevier 85 32 10 6
Springer 30 17 8 3
MDPI 38 26 11 3
Hindawi 18 10 2 1
ACM 8 5 1 1
Others (from
Google
Scholars)
36 23 13 6
Total 340 211 77 30
Path Planning methods classification
Based on
Type
Global path
planning
Local path
planning
Based on Time
domian
Offline time
domain
Online time
domain
2D environment
3D
environment
Based on Space
domain
Map-
based
Potential field
based
Mathematical
based
Evolutionary
based
Based on
Optimization
methods
Conventional
algorithms
Cell-based
algorithms
Model-based
algorithms
Learning-based
algorithms
Based on
ALgorithm
Fig. 1 Path planning methods’ classification
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 6 of 28
Vol:. (1234567890)
organized in two tables for each class of the meta-
heuristic algorithms, showing the experimental details
used and the advantages and limitations of each study.
The experimental table includes the following details:
1. Space domain
2. Time domain
3. Number of iterations
4. Type of obstacles
5. Number of obstacles
6. Environment
7. Population size
8. Application field
The results table includes the following:
1. Proposed algorithm’s name
2. Name of the basic algorithm
3. Algorithm parameters
4. Advantages
5. Limitations of each study
Results andanalysis
Results
In the selection criteria phase, 30 researches have been
retrieved, and each article proposes a new meta-heuristic
algorithm. As mentioned previously, meta-heuristic
algorithms are classified based on different criteria. In
this review, we classify the meta-heuristic algorithms
as evolutionary algorithms, physics-based algorithms,
swarm-based algorithms, and hybrid meta-heuristic
algorithms. Hence, based on this classification, the
retrieved researches were categorized, and the common
factor between all the researches is optimizing path
planning in UAVs using meta-heuristic algorithms.
Figure2 shows the retrieved article distribution based
on the publishing years, from 2010 to 2021. The graph
shows that since 2018 there is a remarkable increase
in the development of novel meta-heuristic algorithms
in the field of UAV path planning. Most of the 2020
and 2021’s papers are novel works on swarm-based
algorithms, evolutionary algorithms, and hybrid
algorithms that mixed swarm-based algorithms with
evolutionary algorithms. This show that these two
class of meta-heuristic algorithm have good results in
optimizing many global engineering problems.
Most of the retrieved researches have been pub-
lished in IEEE (33%). The others are published by
Elsevier, Springer, MDPI, and others by 20%, 10%,
10%, and 26.67% for each publisher, respectively, as
shown in Fig.3.
Among the selected researches, 56% of the studies
are written by authors from China and then 10% from
India, and the other percentages are from other coun-
tries as shown in Figs.4 and 5.
Analysis
The data extracted and collected previously have been
aggregated for answering the research questions. In
the below sections, each research question has been
answered based on the results of the data extraction
process.
RQ1: what are thetypes ofmeta‑heuristic
algorithms?
Figure6 shows the distribution of retrieved researches
based on the algorithm classification; as can be seen,
the swarm-based algorithms are the most used algo-
rithms followed by evolutionary algorithms, hybrid
algorithms, and physic-based algorithms. Table 2
shows the selected researches grouped by the algo-
rithms that are based on.
RQ2: inwhat applications are these algorithms
tested?
As UAVs are used in different areas and for differ-
ent purposes, in the military or civil application,
based on the selected researches, the application
field is varied and primarily applied in military and
battlefield environments. Figure 7 shows the bar
chart for the application fields; as can be seen, the
military has higher study and prospecting the low-
est. The second highest is in general fields. The
researches that not mention the area were consid-
ered as general-purpose algorithms.
RQ3: does theUAV operates ina2D or3D
environment?
Most of them use the 3-dimensional space domain (24)
researches, and 4 researches used the 2-dimensional
space domain; only two researches simulated the
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 7 of 28 30
Vol.: (0123456789)
Fig. 2 Article retrieval
process
Start
Boolean search based on the keywords
First keywords: UAV path planning
Second keywords: UAV path planning
optimisation
Third keywords: meta-heuristic algorithm for
UAV path planning
Selecting articles based on the keywords:
340 articles
Implementing inclusion and
exclusion criteria
30 articles 310 articles
End
Accepted
Rejected
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 8 of 28
Vol:. (1234567890)
Fig. 3 Distribution of
researches based on the
publishing year
Fig. 4 Distribution of
researches based on the
publisher
Fig. 5 Distribution of stud-
ies based on countries
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 9 of 28 30
Vol.: (0123456789)
Fig. 6 Top countries
in UAV path planning
researches using meta-
heuristics on a world map
Table 2 List of selected researches, grouped by the classification of meta-heuristic algorithms
Evolutionary algorithm Adhikari etal. (2017), Oz etal. (2013), Tao and Zheng Wang, (2014), Roberge etal. (2018), Ghambari
etal. (2019), Du etal. (2019), Liu etal. (2019), Yu etal. (2021)
Physic-based algorithm Turker etal. (2015), Wang etal. (2018), Kumar etal. (2018), Huo etal. (2020), Jarray and Bouallègue
(2020), Tang etal. (2021)
Swarm-based algorithm Zhang etal. (2016), Pandey etal. (2018), Dewangan etal. (2019), Fan and Akhter (2021), Yang etal.
(2020), Ge etal. (2020), Shao etal. (2020), Shin and Bang (2020), Zhou etal. (2021)
Hybrid algorithm Duan etal. (2010), Wang etal. (2012), Qu etal. (2020b),Qu etal. (2020a), Biundini etal. (2021), Da Silva
Arantes etal. (2016), Pan etal. (2020)
Fig. 7 Distribution of
researches based on the
algorithm classification
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 10 of 28
Vol:. (1234567890)
proposed algorithm in both 2- and 3-dimensional space
domain, as shown in Fig.8.
RQ4: does theUAV operates inoffline oronline time
domain?
All the retrieved researches used the offline time
domain, which is due to the difficulty in practicing
the novel algorithms in the online time domain. Only
the article (Du etal., 2019) tested the proposed algo-
rithm in both online and offline environments, as
shown in Fig.9.
RQ5: does UAV use maps orsensors toreach
thedestination?
Most of the researches used simulation environment
for creating the maps for the UAV path flights. The
most used simulation is MATLAB, and around 68%
of the authors used this tool. Figure10 shows in detail
all the simulation applications that have been used.
RQ6: what are thetype ofobstacles have been used
intheenvironment andtheir numbers?
One of the challenges in UAV path planning is hav-
ing obstacles in the flight environments. Obstacles are
Fig. 8 Classification based
on application fields
Fig. 9 Distribution of researches based on the environment
dimensions Fig. 10 Distribution of researches based on the time domain
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 11 of 28 30
Vol.: (0123456789)
something that blocks the movement, going ahead, tak-
ing action, or progress. The UAVs must avoid these
obstacles to reach the destination with optimal or near-
optimal solutions. Most of the researches that proposed
algorithms for general-purpose applications used dif-
ferent obstacles, such as cylinder, circle, and sphere
obstacles. At the same time, the military and battlefield
applications mainly used radar and flying objects, espe-
cially flying UAVs. Figure11 shows the classifications.
The number of obstacles varies from one to two if the
obstacles are mountains to up to 80 obstacles for the
cylinder and circle obstacles.
RQ7: does thealgorithm tested onasingle UAV
ormultiple UAVs?
UAVs’ number in the simulation environment is the
final question, because the number of UAVs affected
the results and the algorithm implementations. Fig-
ure12 below shows the number of UAVs, and most
Fig. 11 Analysis of the
simulation environment
15
12
1111
0
2
4
6
8
10
12
14
16
Simulaon applicaons for mapping
Fig. 12 Analysis of the
type of obstacles
2
8
232
13
0
2
4
6
8
10
12
14
Terrain
obstacles
Mountain RadarFlying
objects
Balefield
threats
Different
shape
obstacles
Obstacles
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 12 of 28
Vol:. (1234567890)
of the studies used a single UAV for implementing
the proposed algorithms (Fig.13).
Discussion
In general, path planning in robotics refers to the pro-
cess of traveling from one point (start point) to another
(destination point) while avoiding obstacles in order to
reach the required least cost. Path planning considers
as a primary issue in robotics in general and in UAVs
in specific. Their difficulty changes with their mis-
sions and the mission environment. The UAVs used in
the military and on the battlefields differ from those
used in surveillance, rescue missions, and detecting
fires in forests. In each field, they use different tech-
niques such as in forest fire detection; they deployed
UAVs to cover broad areas in both daytime and night
with a long-distance and even in cloudy weather.
Besides, using UAVs for detecting fire needs less time,
and it is inexpensive. UAVs use infrared (IR) cameras
for forest fire detection to detect hot objects that are
represented as bright areas and detect smoke motion
(Sudhakar etal., 2020; Yuan etal., 2017). As the use
of UAVs rises, their missions get more complex, and
as a result, path planning is getting more complex.
Although many research works for optimizing path
planning in UAVs using different ways, path planning
optimization is still a broad area for study (Tsourdos
et al., 2010). All existing techniques, methods, and
algorithms aim to find optimal or near-optimal paths
using optimization methods by considering the obsta-
cles and local conditions such as winds.
Modern problems have become too complicated
for old approaches to handle, and meta-heuristic
algorithms are considered the most effective alter-
native solutions (Ri et al., n.d.). Mostly, the meta-
heuristic algorithms are nature-inspired and have
many advantages over the other algorithms and meth-
ods such as these algorithms can solve a complex
problem without the need for gradient information and
can be easily implemented. Meta-heuristic algorithms
apply to different problems. Finally, they can solve
the local optima problem that most other algorithms
fall into (Mirjalili & Lewis, 2016; Khalilpourazari &
Khalilpourazary, 2019). However, there is no guaran-
tee that the given and suggested solutions will solve
the existing problems (Gandomi etal., 2013).
Meta-heuristic algorithms are classified into dif-
ferent forms based on their focus and characteristics
(Gandomi etal., 2013). It can be classified based on
the gradient, based on the trajectory, based on the
search capability, and many other classifications.
In this review, we classified meta-heuristics based
on the inspiration into evolutionary algorithms,
physics-based algorithms, swarm-based algorithms, and
hybrid-meta-heuristic algorithms (Mirjalili & Lewis,
2016; Gandomi etal., 2013). In each class, algorithms
consider the overall length of the flight, terrain and
obstacles, weather, and threat factors. Besides differ-
ent environments, the environment map is either 2D
or 3D, and path planning is local or global.
Fig. 13 Analysis of the
number of used UAVs
21
9
0
5
10
15
20
25
Single UAV Mulple UAVs
Number of UAVs
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 13 of 28 30
Vol.: (0123456789)
First: evolutionary algorithms
Evolutionary algorithms (EA) are a probabilistic opti-
mization technique influenced by theoretical organic
evolution and based on collective learning among
individuals. Each is a starting point for a search
throughout the universe of possible solutions to a spe-
cific problem using various strategies and procedures
(Fortin et al., 2012; Bäck & Schwefel, 1993; Beyer
etal., 2002). EA has recently become a popular and
widely used optimization tool. Evolutionary algo-
rithms returned in the late 1960s; first, Schwefel, and
Rechenberg introduced the evolutionary strategies;
then, in the 1970s, Holland introduced the genetic
algorithm (GA), followed by evolutionary program-
ming by Fogel (Zelinka, 2015). Since then, many
researchers have introduced novel evolutionary algo-
rithms for optimization in different fields. Genetic
algorithm (GA), evolution strategy, genetic program-
ming, evolutionary programming, and many other
evolutionary algorithms are among the most common.
Differential evolution (DE) is a unique heuristic
technique for reducing nonlinear and non-differentiable
continuous space functions. It converges more rapidly
and precisely than several other well-known techniques
to global population-based optimization. This method
makes DE more resilient, straightforward to implement,
and well-suited to parallel computation [53][54]. Using
differences between randomly chosen alternative solu-
tions, DE changes successive approximations of solu-
tions. Mutations have significant amplitudes when can-
didate solutions are selected widely. Mutations are less
amplitudes if candidate solutions are chosen in a con-
fined area (Gavrilas, 2010). Adhikari et al. (2017) and
Yu etal. (2021) proposed new algorithms based on the
differential evolution for optimizing the UAV path plan-
ning, and compared with other algorithms, the simula-
tion results show that the algorithm proposed in Adhikari
etal. (2017) has improvement rate of 4 to 8%, while Yu
et al. (2021) proposed an algorithm that can achieve a
smooth path to guide the UAV, and it has less computa-
tional time compared to other algorithms. However, they
are not applicable with dynamic obstacles and complex
environments.
A genetic algorithm (GA) successfully addresses
the path-planning problem and contributes to the
realization of autonomous navigation and control for
unmanned surface vehicles. GAs are search strate-
gies based on specific genetics and natural selection
mechanisms, using three basic operators: selection,
crossover, and mutation. GA is based on genetics
and natural selection mechanisms, using three basic
operators: natural selection, crossover, and mutation.
The GA strategy can be summarized as follows: first,
the chromosomes are randomly selected around the
search space to produce new offspring, based on the
fitness function. Chromosomes with better fitness
are chosen. Then the chosen chromosomes enter the
crossover stage to generate two offspring from both
parents with inherent characteristics from both of
them. The mutation process will alter the newly cre-
ated individual and will be altered by the mutation
process to ensure novelty after completing the off-
spring population; they will be replaced by the pre-
vious parents (Gavrilas, 2010; Malik & Tayal, 2014;
Kaveh & Farhoudi, 2011). GA has been applied to
various types of autonomous equipment to solve tra-
jectory planning problems due to its increased resil-
ience and more robust global search capability (Xin
et al., 2019). Then creating new generations after
each iteration and the new chromosome will be evalu-
ated for creating another new generation. The itera-
tion will be repeated until the best chromosome is
created as an optimal solution to the problem (Said
etal., 2014).As shown in Tables3 and 4, Oz etal.
(2013), Zekui et al. (2018), Roberge et al. (2018),
and Ghambari et al. (2019), are proposed new algo-
rithms based on the genetic algorithm that was tested
in offline and 3D static environments with different
types and numbers of obstacles with different appli-
cation fields including terrain and restricted area and
military and general-purpose applications, although
the experimental results of these proposed algorithms
show an improvement to get a near-optimal solution,
shorter paths, lower cost, and better performance.
Hence, the proposed algorithms suffer from long
execution times, applied only on static obstacles, not
dynamic ones, and the terrain environments still do
not solve the UAV path planning problem.
Based on the evolutionary algorithm and some
PSO features, [29] proposed an algorithm that has
good performance in the simulation environment.
But, the time required for obtaining the near-optimal
solution is very long and longer than the algo-
rithm that has been compared. Finally, based on the
improved t-distribution, Liu et al. (2019) proposed
an evolutionary optimization algorithm with good
results in dynamic path planning with complex and
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 14 of 28
Vol:. (1234567890)
unknown geographic information. However, the algo-
rithm can optimize the path only if actual data about
the environment and the constraint conditions are
available (Tables5, 6, 7, 8, 9, and 10).
Second: physic-based algorithms
As many scientists employ physical, chemical, and
biological laws to improve novel optimization meth-
ods to solve complex, multimodal, high-dimensional,
and nonlinear problems, new optimization meth-
ods are being developed. The concept of physics-
based algorithms was proposed, and various scien-
tists predicted that these algorithms would be used
to investigate the related literature [35] thoroughly.
Physic-based algorithms have many algorithms
such as simulated annealing, Multi-Verse Optimizer
(MVO), Gravitational Search Algorithm, and Central
Force Optimization Algorithm.
To begin, simulated annealing (SA) is an optimi-
zation technique that can control cost functions and
arbitrary boundaries of nonlinearities, discontinui-
ties, and stochasticity (de Sales Guerra Tsuzuki etal.,
2006). SA was developed on the Metropolis method
and is based on moving in the search space towards
states with lower fitness function values (Mirjalili
etal., 2016). Based on the classic simulated annealing,
Turker etal. (2015), Wang etal. (2018), and Huo etal.
(2020) are proposing new algorithm based on simu-
lated annealing (SA). Turker etal. (2015) simulated
Table 3 Experimental details for evolutionary algorithms
Ref Space
domain Time
domain Number of
iterations Type of
obstacles Number of
obstacles Environment
and obstacles Population
size Application
field
Adhikari
etal.
(2017)
3D Offline 100 Sphere zones 8 Static 15 General
Oz etal.
(2013)
3D Offline 20 Smooth
terrains
and rough
terrains
4 Static 200 Rescue
missions
in terrain
environ-
ment
Zekui etal.
(2018)
3D Offline 20 Mountains 7 Static – Restricted
area
Roberge
etal.
(2018)
3D Offline 100 No-fly zone
as cylindri-
cal shape
with an
infinite
height
4 Static 32 Military
Ghambari
etal.
(2019)
2D and 3D Offline 20 Using stand-
ard circles
in 2D envi-
ronment
and simple
cuboidal in
3D envi-
ronment
80 in 2D
and 33 in
3D
Static 100 General
Du etal.
(2019)
3D Offline and
Online
100 Mountains Terrain
obstacles
(such as
rivers and
cliffs)
Static 60 and 45 Rescue
operations
Liu etal.
(2019)
3D Offline 250 Mountains Static General
Yu etal.
(2021)
3D Offline 30 Mountains – Static – Rescue mis-
sions/disas-
ter area
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 15 of 28 30
Vol.: (0123456789)
Table 4 List of the proposed algorithm, algorithm parameters, results, and algorithm limitations
Ref Proposed algorithm Based algorithm Algorithm parameters Advantages Limitations
Adhikari etal. (2017) Fuzzy logic adaptive differ-
ential evolution (FA-DE)
Differential evolution Differential weight, crosso-
ver rate, and mutation
operator
The proposed algorithm has
a better performance by
4% and 8% compared to
DE/best/1 and DE/rand/1,
respectively
Not applicable on dynamic
obstacles
Oz etal. (2013) Two algorithms based on
GA and HH
Genetic algorithm and
hyper-heuristic
Population size, crossover
rate, and mutation rate
In the smooth terrain, the
GA algorithm achieved
the lowest cost values.
While in the rough ter-
rain, the HH algorithm
achieved better results
Tao and Zheng Wang
(2014)
Multi-objective genetic
algorithm
Genetic algorithm Travel ability, diversity,
crossover operator, and
reconstruction operator
The crossover operator gets
high in conjunction with
the population, and when
the path is found, the
reconstruction operator
gets a high chance to be
applied
Weak in terrain processes
Roberge etal. (2018) Parallel implementation of
the genetic algorithm
Genetic algorithm Fitness, runtime, and
speedup
Reaching in near optimal
path in a short time
Although detecting obstacles
in a long-distance could
be accepted theoretically,
implementing the same
idea on CPU takes a long
time as 290 times longer
than the theoretical execu-
tion time
Ghambari etal. (2019)Improves teaching learning-
based optimization
(TLBO)
Genetic algorithm Crossover rate, mutation
rate, and the maximum
number of runs
Comparing to the standard
TLBO, the proposed algo-
rithm has better perfor-
mance in both 2D and 3D
environments
Dynamic obstacles and the
efficiency of the algorithm
with multiple UAVs
Du etal. (2019) Hybrid evolutionary algo-
rithm
Evolutionary algorithm
and PSO
Migration, mutation, and
local search
The proposed method
presents a good perfor-
mance, and it can provide
a notable improvement in
real life
Producing solution in a long
time, longer than the com-
pared algorithm
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 16 of 28
Vol:. (1234567890)
the algorithm in military environment with radar
as obstacles, and they found that the performance
increases with the increase of the iteration time.
However, this increases in number of iterations and
increases the elapsed time. Simultaneously, Wang etal.
(2018) and Huo et al. (2020) proposed novel algo-
rithms, and the results of them show that they have
good performance with acceptable computational time
but are not applied on complex environments espe-
cially when the UAV is used for rescue missions.
The second algorithm of physic-based algorithms
is Multi-Verse Optimizer that is mainly inspired by the
three concepts in cosmology: white hole, black hole,
and wormhole, with mathematical models to perform
exploration, exploitation, and local search, respec-
tively. According to this algorithm, the probability of
having white holes is directly related to the inflation
rate, and individuals with a higher inflation rate have
a greater probability of sending objects through white
holes. At the same time, individuals with a lower
inflation rate are more likely to send items through
black holes. The objects in all universes are randomly
transported through wormholes to the current best
individual (universe). This random process occurs
regardless of the inflation rate. The MVO algorithm
has been applied on path planning problem as well as
applied to many real engineering problems (Mirjalili
etal., 2016). Many researchers used the MVO algo-
rithm and improved it for optimization; among them,
Jarray and Bouallègue (2020) proposed the Multi-
objective Multi-Verse Optimizer (MOMVO) and com-
pared it with 4 other meta-heuristic algorithms, and
the proposed algorithm has better performance than
the other algorithms. But the experimental was only
on the static environments.
Further, Kumar etal. (2018) proposed a new tech-
nique for finding a suitable optimizer among six
meta-heuristic algorithms (ALO, DA, GWO, MFO,
WOA, and MVO). They used flying UAVs and fly-
ing objects like birds as threats in a 3D environment,
and the proposed techniques show that the MVO
has the least average fitness function value. Finally,
Tang etal. (2021) proposed a new algorithm based
on Equilibrium Optimizer (EO) in both 2D and 3D
offline environment with different static cylindrical
obstacles. The experimental results show that in a no-
threat environment, the proposed algorithm has good
results with short time, but in threat environments, it
takes a longer time for finding the path.
Table 4 (continued)
Ref Proposed algorithm Based algorithm Algorithm parameters Advantages Limitations
Liu etal. (2019) Evolutionary optimization
algorithm
The improved t-distribution In dynamic track plan-
ning with complex and
uncertain geographic
knowledge, the method is
reasonable
Only if there is actual data
about the surroundings and
the constraint conditions,
then the path is optimized
Yu etal. (2021) A knee-guided differen-
tial evolution algorithm
(DEAKP)
Differential evolution
algorithm
Turning angle and safe
distance
Generation of UAV’s
smooth path with a
computational time of less
than 200s
Not applicable with dynamic
obstacles and complex
environments
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 17 of 28 30
Vol.: (0123456789)
Third: swarm-based algorithms
The swarm-based algorithm is a feature of a system
in which the combined behaviors of unsophisticated
individuals interacting locally, which creates a coher-
ent functional global pattern to develop that is known
as swarm intelligence (Saka etal., 2013). The swarm-
based algorithms are mostly inspired by bird colonies,
ants, schools of fish, and herds of animals (Game,
2020). The problem-solving approaches of swarm
intelligence have various advantages over more tra-
ditional methods. They are simple, robust, and give
a solution without centralized control or the deploy-
ment of a global model.
The swarm-based algorithm has many algorithms
based on swarm behavior, such as grey wolf optimizer
(GWO). GWO is a novel meta-heuristic optimization
method. Its guiding premise is to mimic the coop-
erative hunting behavior of grey wolves in nature. In
terms of model structure, GWO is distinct from oth-
ers, and it is a large-scale search approach (Niu etal.,
2019). This algorithm divided the wolves in the social
hierarchy into alpha, beta, omega, and delta wolves,
each with their responsibilities. Grey wolves are hunt-
ing in a group, and the first is tracking, chasing, and
approaching the prey. Then, they pursue, encircle, and
harass the prey until it stops moving, and finally, they
attack the prey (Mirjalili etal., 2014).
Based on GWO, Zhang et al. (2016), Dewangan
et al. (2019), and Yang et al. (2020) are proposing
new meta-heuristic algorithms. A new meta-heuristic
grey wolf optimizer (GWO) based on GWO to find
an optimal or near-optimal flight path in light of the
dangers and limits in the combat zone. The proposed
algorithm was compared to CS, FPA, NBA, BSA,
ABC, and GGSA algorithms. The results show that
the algorithm is very effective and has a high level of
local optimal avoidance. Furthermore, the accuracy
of the obtained optimal values for weighted sum cost
is perfect, owing to the GWO’s extensive use. Also,
Dewangan etal. (2019) proposed a new algorithm as
improved grey wolf optimizer (GWO) and compared
it with Dijkstra, A* and D* (deterministic), BBO,
IBA, PSO, GSO, WOA, and SCA algorithms; the
results show that the proposed model has less cost
time and less distance compared to other algorithms.
Finally, Yang etal. (2020) proposed a new algorithm,
multi-population chaotic grey wolf optimizer (MP-
CGWO), with the aim of finding multi-UAV coop-
erative path planning; the minimum cost of proposed
Table 5 Experimental details for physic-based algorithms
Ref Space
domain Time
domain Number of
iterations Type of
obstacles Number of
obstacles Environ‑
ment and
obstacles
Popula‑
tion
number
Application
field
Turker etal.
(2015)
2D Offline 921,030 Regular and
circular
radars
10 Static – Military
Wang etal.
(2018)
3D Offline 200 Terrains–
mountains
– Static – Rescue mis-
sions from
the forest
fire
Kumar etal.
(2018)
3D Offline 1000 Flying
objects
like other
UAVS,
birds, or
aircrafts
2 Dynamic 30 General
Huo etal.
(2020)
3D Offline 600 – Small Num-
ber
Static 50 General
Jarray and
Bouallègue,
(2020)
3D Offline 100 Cylinder-
shaped
obstacles
16 Static 50 General
Tang etal.
(2021)
2D and 3D Offline 600 Cylindrical
obstacles
6 Static 500 General
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 18 of 28
Vol:. (1234567890)
Table 6 List of the proposed algorithm, algorithm parameters, results, and algorithm limitations
Ref Proposed algorithm Based algorithm Algorithm parameters Advantages Limitations
Turker etal. (2015) Simulated annealing (SA)
algorithm
Simulated annealing
algorithm
Cooling rate, distance
energy, threat radius, and
radius value
By the number of iterations
(88), the improved
percentage of the
performance is 36%,
while the improved
percentage of the
performance is 82% when
the number of iterations
is 921,030. Thus, the
performance increases
with the increase in the
number of iterations
Along with the increase
in iteration numbers, the
elapsed time increases
Wang etal. (2018) Vortex Search (VS)
algorithm
Based on SA (simulated
annealing), PS (pattern
search), and population-
based algorithms
Candidate solution number
and maximum iteration
The VS algorithm takes
0.2s as computational
time, while PSO takes
5.2s
Only mountains are taking
into account as threats. The
fire forest environment has
many threats and dynamic
Kumar etal. (2018) A technique for finding
suitable optimizer among
these meta-heuristic
algorithms: Ant Lion
Optimizer (ALO),
Dragonfly Algorithm
(DA), grey wolf optimizer
(GWO), Moth Flame
Optimization (MFO),
Whale Optimization
Algorithm (WOA), and
Multi-Verse Optimizer
(MVO)
– Fitness function value and
computation time
MOV has the smallest
average fitness function
of 0.152 and the shortest
average computation time
of 33.686s
–
Huo etal. (2020) Swap-and-Judge Simulated
Annealing (SJSA)
algorithm
Simulated annealing Runtime SJSA outperforms DIDE,
MGA, and DISA by
approximately 38.17%,
37.10%, and 30.26%,
respectively. Besides, it is
more efficient at finding
near-optimal solutions
than the exact algorithm
CPLEX
Not applicable in dynamic
and complex environments
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 19 of 28 30
Vol.: (0123456789)
algorithm is almost close to the compared algorithms,
(GWO, PSO, and DE), and its stability is improved
by 56%.
Another type of swarm-based algorithm is the
glowworm swarm optimization (GSO) algorithm,
which is a novel meta-heuristic algorithm based on
the behavior of glowworms. It can capture all the
maximum multimodal functions. The idea of this
algorithm is from the biological behavior of glow-
worms that can emit light and use the biolumines-
cence glow for different purposes (Gong etal., 2011;
Ludwig, 2016). Under the many advantages of the
GSO, many researchers used this algorithm for opti-
mizing the UAV path planning; Pandey etal. (2018)
proposed a modified and improved GSO, and it has
been tested in a static terrain 3D environment with
comparison to other algorithm such as Dijkstra, PSO,
IBA, and BBO algorithms. in static 3D environ-
ments. The results show that if there is a clear path
from source to destination, the proposed algorithm
will cost less for finding the path. The not clear path
from source to destination makes the performance
of this algorithm less; besides, with the increase of
population size, the performance decreases. These
two are the most obvious limitation of the proposed
algorithm.
Particle swarm optimization is a swarm-based
method based on a simplified social model closely
related to swarming theory introduced by Kennedy
and Eberhart. For instance, a swarm of bees searches
for a food source where each bee uses its memory and
knowledge gained by the swarm as a whole to find
the best available food source. The PSO is simple to
implement since it requires a small number of param-
eters. PSO is based on a population of particles that
hover through the problem search space through the
interaction of individual particles. The particles then
update their positions based on the local and global
best solutions. If the best local solution has a cost less
than the cost of the current global solution, then the
best local solution replaces the best global solution
(Malik & Tayal, 2014; Gavrilas, 2010; El-henawy &
Abdelmegeed, 2018).
Based on the swarm ideas, Fan and Akhter (2021)
proposed a time-varying adaptive inertia weight
parameter for PSO called nPSO for UAV path plan-
ning optimization with the aim of avoiding falling into
local minima. The proposed nPSO algorithm includes
sw-nPSO, tw-nPSO, aw-nPSO, aw2-nPSO, and
Table 6 (continued)
Ref Proposed algorithm Based algorithm Algorithm parameters Advantages Limitations
Jarray and Bouallègue
(2020)
Multi-objective Multi-Verse
Optimizer (MOMVO)
Multi-Verse Optimizer
(MVO)
MOMVO algorithm
outperformed the
compared algorithms
(MSSA, MOGWO, and
NSGA-II) for over 21%,
35%, and 1%, respectively
Dynamic obstacles
Tang etal. (2021) Multiple Population Hybrid
Equilibrium Optimizer
(MHEO)
Equilibrium Optimizer
(EO)
Battlefield space, number
of path nodes, maximum
turn angle, and the
maximum climb angle
In a no-threat environment
and compared to EO,
MHEO has better results
with less time. While in
threat environments and
comparing to MPA and
GEDGWo algorithms,
MHEO has a better mean,
but it takes a longer time
for finding the path
Longer time compared with
other algorithms in threat
environments
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 20 of 28
Vol:. (1234567890)
taw-nPSO. Comparing the proposed nPSO, the results
show that the taw-nPSO has a better outcome with
less cost when the number of iteration increases. But
it is not applicable with multivariable inertia weight.
With the same idea of the PSO and for optimizing
UAV path planning in battlefield environment with
dynamic obstacles, Shin and Bang (2020) proposed
a new and improved PSO algorithm. Shin and Bang (2020)
proposed a new and improved PSO algorithm. After
several iterations and different numbers of particles, and
comparing it with standard PSO, it has better perfor-
mance in both short-range and long-range paths while
satisfying all dynamic restrictions. While Shao et al.
(2020) proposed comprehensively improved particle
swarm optimization (CIPSO) that is based on PSO,
the results show that there are improvements in the
Table 7 Experimental details for swarm-based algorithms
Ref Space
domain Time
domain Number of
iterations Type of
obstacles Number of
obstacles Environment
and obstacles Population
size Application
field
Zhang etal.
(2016)
2D Offline 30 Threats in
battlefield
environ-
ment
21 Static 200 Military-
combat
field
Pandey etal.
(2018)
3D Offline 25–50 Cuboids of
different
magnitudes
14 Static 20–30 Terrain envi-
ronment
Dewangan
etal.
(2019)
3D Offline 25–50 Static obsta-
cles with-
out more
information
about them
16 Static 20–30 General
Fan and
Akhter
(2021)
2D Offline 100 Circular-
shaped
static
obstacles
15 Static 100 General
Yang etal.
(2020)
3D Offline 150 Mountain
peaks
6 Static 150 Battlefield
Ge etal.
(2020)
3D Offline 500 Birds, vehi-
cles, and
oil tank
trucks are
examples
of mobile
obstacles.
Static
impedi-
ments
include
trees, oil
tanks, and
pumping
wells
42 Dynamic 80 Oilfield
inspections
Shao etal.
(2020)
3D Offline 120 Mountains
and radar
2 mountains
and 1 radar
Static 500 General
Shin and
Bang
(2020)
3D Offline 1000 SAM–radar 20 Dynamic 40 swarms
and 2000
particles
Military and
battlefield
Zhou etal.
(2021)
3D Offline 25–50 Different-
shaped
static
obstacles
26 Static 50 Military
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 21 of 28 30
Vol.: (0123456789)
Table 8 List of proposed algorithm, algorithm parameters, results, and algorithm limitations
Ref Proposed algorithm Based algorithm Algorithm parameters Advantages Limitations
Zhang etal. (2016) Meta-heuristic grey wolf
optimizer (GWO)
Grey wolf optimizer (GWO) a It avoids local optima at
a very high level, which
increases the probability
of getting a good
approximation to this
path’s optimal weighted
sum cost
The accuracy has a very high
weighted sum cost
Pandey etal. (2018) Improved glowworm swarm
optimization (GSO)
Glowworm swarm
optimization (GSO)
Population size, number
of iterations, initial
luciferin, initial range,
luciferin decay, luciferin
enhancement, range
boundary, and range
constant
With big maps, the
improved GSO has
significant movement.
Additionally, it is capable
of avoiding local optima.
The new technique
demonstrates a promising
solution for real-time UAV
pathfinding in the static
impediments
The proposed algorithm
slows down with a large
population size
Dewangan etal. (2019) Improved grey wolf
optimizer (GWO)
Grey wolf optimizer (GWO) Parameter (a), population
size and maximum
iterations
The improved algorithm
finds an optimized path
with static obstacles and
lower path calculation than
other meta-heuristic and
deterministic algorithms. It
uses fewer parameters that
lead to improvement in the
speed of convergence
–
Fan and Akhter (2021) nPSO including: sw-nPSO,
tw-nPSO, aw-nPSO, aw2-
nPSO, and taw-nPSO
PSO Weigh, population size, and
iteration number
The taw-nPSO has better
outcomes compared
with the other proposed
algorithms
Not applicable on
multivariable inertia weight
Yang etal. (2020) Multi-population chaotic
grey wolf optimizer
(MP-CGOW)
Grey wolf optimizer Number of populations,
iterations, and number of
waypoints
The stability is improved by
(56%) compared to other
algorithms
Compared to the original
GWO, the planning time
remains with no changes
when the complexity of the
algorithm increases
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 22 of 28
Vol:. (1234567890)
Table 8 (continued)
Ref Proposed algorithm Based algorithm Algorithm parameters Advantages Limitations
Ge etal. (2020) Improved pigeon-inspired
optimization algorithm
(PIOFOA)
Pigeon-inspired optimization
(PIO) algorithm and fruit
fly optimization algorithm
(FOA)
The space dimension,
population size, map, and
compass operator
The suggested PIOFOA
method has a reasonable
computation time but not
least among the comparing
algorithm and a proper
cost function. Besides,
PIOFOA is more efficient
in environments with
moving obstacles
Long computational time
Shao etal. (2020) Comprehensively
improved particle swarm
optimization (CIPSO)
PSO Average fitness value,
average minimum
iteration, and path failure
rate
The improved percentage of
the average fitness value
is 22.69%, the improved
percentage of the average
minimum iteration is
33.11%, and the improved
percentage of the path
failure rate is 85.71%
High total path failures
Shin and Bang (2020) Improved particle swarm
optimizer (PSO)
PSO Particle dimension, number
of particles, constriction
factor, individual learning,
and global learning eco-
efficiency
The performance for
short-range with a single
swarm is 0.93, and it is
0.92 for a long-range
test. At the same time,
the performances with 10
swarms have higher results
and become 0.97 and 0.96
–
Zhou etal. (2021) Improved bat algorithm
integrated into the ABC
algorithm (IBA)
PSO and BA Crossover and mutation IBA can swiftly arrange a
battle path for UAVs that
avoids mountains and
dangerous threats. Besides,
IBA outperforms DE,
BAM/ABC/PSO/IABC/
GFACO in the battlefield
environment with a slower
convergence rate and less
convergence time
Not working in dynamic
environments
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 23 of 28 30
Vol.: (0123456789)
percentage of the average fitness value, the percentage of
the average minimum iteration, and the percentage of the
path failure rate. Also, Zhou etal. (2021) proposed the
improved bat algorithm integrated into the ABC algo-
rithm (IBA) that are based on PSO and BA. Simulations
show that IBA can swiftly arrange a battle path for UAVs
that avoids mountains and dangerous threats. Further-
more, the comparison results show that the IBA outper-
forms DE and BAM/ABC/PSO/IABC/GFACO in the
battlefield environment with a slower convergence rate
and less convergence time. Although the proposed algo-
rithm has some improvement compared to the other algo-
rithms, it has not been tested in the dynamic environment
and the battlefield is mostly a dynamic environment.
Finally, Ge et al. (2020) proposed an improved
pigeon-inspired optimization algorithm (PIOFOA)
that is based on pigeon-inspired optimization (PIO)
algorithm and fruit fly optimization algorithm (FOA)
for solving the path planning in dynamic 3D oilfield
inspections environment. The suggested PIOFOA
method is efficient in dynamic environments. How-
ever, the computational time is high, and there is
not much improvement comparing with the standard
algorithms that the proposed algorithms are based on.
Fourth: hybrid meta-heuristic algorithms
Recently, many researchers have found that several
algorithms do not follow a single traditional meta-
heuristics paradigm. That leads to new ideas by com-
bining different algorithms, especially the use of
local search methods in population-based methods.
Hybridization has different types such as hybridiza-
tion between meta-heuristic algorithms and known
as hybrid meta-heuristic algorithms or hybridizing
with constraint programming or with tree search tech-
niques (Blum etal., 2015).
For finding the shortest path for a UVA in mili-
tary applications and battlefield environments, Duan
Table 9 Experimental details for hybrid meta-heuristic algorithms
Ref Space
domain Time
domain Number of
iterations Type of
obstacles Number of
obstacles Environment
and obstacles Population
size Application
field
Duan etal.
(2010)
3D Offline 15 Sphere
threaten
area
4 Static – Military
Wang etal.
(2012)
3D Offline 100 Threats in
enemy
environ-
ment
3 Static 15—40 Military
Qu etal.
(2020b)
3D Offline 500 Circular
obstacles
8 Static 50 General
Qu etal.
(2020a)
3D Offline 1000 Cylinder
obstacles
8 Static – Complex flight
environment
Biundini
etal.
(2021)
3D Online and
offline
200 – – Static 50 General–
terrain envi-
ronment
Da Silva
Arantes
etal.
(2016)
2D Offline 100 Non-convex
environ-
ment with
uncertain-
ties by the
presence
of no-y
zones
such as
mountains,
cities, and
airports
20 Static 13 General
Pan etal.
(2020)
– Offline 100 Circular
obstacles
10 Static 60 General
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 24 of 28
Vol:. (1234567890)
Table 10 List of the proposed algorithm, algorithm parameters, results, and algorithm limitations
Ref Proposed algorithm Based algorithm Algorithm parameters Advantages Limitations
Duan etal. (2010) Hybrid meta-heuristic
ant colony optimization
(ACO) and differential
evolution (DE) algorithm
Ant colony optimization
(ACO) and differential
evolution (DE) algorithm
Alpha, beta, p
(pheromone decay
parameter), q, and Q
Optimal path decreased by 2.77%
when the number of UCAVs
is 10, while when the number
increased to 20, the optimal
path decreased to 2.26%. This
shows the improved
The performance is
decreasing, while the
number of UAVs is
increasing
Wang etal. (2012) Hybrid meta-heuristic
differential evolution
(DE) and cuckoo search
(CS) algorithm
Differential evolution and
Cuckoo search algorithm
Step size, population
size, discovery
rate, and maximum
generation Maxygen
It has less time (46.53s)
comparing to the basic CS
algorithm. Therefore, it is
feasible, practical, and flexible
for battlefield environments
The proposed algorithm
has problems with the
fleet formation, target
distribution, arrival time
constraint, and avoidance
conflict
Qu etal. (2020a) Hybrid algorithm called
HSGWO-MSOS
Based on simplified grey
wolf optimizer
(SGWO) and modified
symbiotic organisms
search (MSOS)
The HSGWO-MSOS method
outperforms GWO, SOS, and
SA by successfully acquiring
effective and safe routes
The search effect is poor
Qu etal. (2020a) Reinforcement learning-
based grey wolf optimizer
algorithm (RLGWO)
Grey wolf optimizer Weighting, learning rate,
discount factor, and
punishment
HSGWO-MSOS algorithm can
obtain an effective and safe
path successfully
–
Biundini etal. (2021) Coverage Path Planning
(CPP) suing genetic
algorithm and bat
algorithm
Genetic algorithm and bat
algorithm
Mission time, coverage
vertical, and coverage
horizontal
With the increase of the distance
that point density increased. As
a result, the performance time
(time to perform the mission)
increased as well. Also, the
density points increased with
the increase of the horizontal
coverage and vertical coverage.
As a result, the mission time
increased
Not using any threats and
obstacles
Da Silva Arantes etal.
(2016)
Hybrid genetic algorithm
(HGA)
Combination of genetic
algorithm and visibility
graph
Number of crossovers,
mutation rate, and
stopping criterion
Finding solutions in less than
10s
HGA quality is as the
same as the quality
reached by CPLEX and
CSA
Pan etal. (2020) A modified CIPDE
(MCIPDE) with modified
JADE (MJADE) called
CIJADE
Based on CIPDE and JADE – Compared to PSO, DE, ABC,
JADE, and CIPDE, the CIJADE
algorithm is more efficient for
finding optimal solutions
–
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 25 of 28 30
Vol.: (0123456789)
etal. (2010) and Wang etal. (2012) proposed hybrid
algorithms. The first one combined ant colony opti-
mization (ACO) and differential evolution (DE) algo-
rithm and compared the new hybrid algorithm with
standard ACO. The simulation results show that the
improved ACO has better performance than the basic
ACO, in which the optimal path decreases by 2.77% and
2.36%, while the number of UCAVs is 10 and 20,
respectively, while the proposed hybrid algorithm in
[39] combines differential evolution (DE) and cuckoo
search (CS) algorithm and compares the hybrid algo-
rithm with standard CS. Although the results show
that the time decreased and improved, the proposed
algorithm has problems with the arrival time con-
straint and avoidance conflict. Likewise, [43] com-
bines a genetic algorithm and visibility graph and a
new hybrid algorithm proposed as a hybrid genetic
algorithm (HGA) to solve the path planning in a 2D
environment. In this algorithm, after creating popu-
lations, the algorithm generates new individuals for
each population. Firstly, it creates the individuals ran-
domly, while in the next execution, all the individuals
will be initialized except the best-migrated individual.
By this procedure, the partial solution turns into a full
solution (UAV route) using the linear programming
model. The computational results are compared with
the exact (CPLEX) and heuristic (CSA) methods.
The results show the HGA has robustness for find-
ing quick solutions in a very low time. However, the
results of the new algorithm are quite similar to the
static one, which means that there is no improvement.
Additionally, Biundini etal. (2021) proposed a hybrid
meta-heuristic algorithm based on the GA and BA,
known as Coverage Path Planning (CPP). The aim of
CPP is to find a path that passes through all of the points
in a given area. The CPP is reliant on the development
of a waypoint mission that fits the mission time and pho-
tometry requirements. The task will then be submitted to
the UAV, which will begin flying. The framework per-
forms a local mini-mission along the path, boosting the
picture data for that location, and then the UAV moves
on to the next waypoint. In the case of failure, the plane
will proceed to the next waypoint. If the waypoint is the
final of the horizontal movement, the UAV examines if
it is the end of the global mission. If this is not the case,
the mission is re-optimized to reduce the impact of local
missions and flight delays. The UAV mission will be
completed if the point is the end one, and the UAV will
return to its launch point.
The simulation results show that the point density
increases with the increase of the distance, horizontal
coverage, and vertical coverage. These increase the
performance time (time to perform the mission) and
increase the mission time. The proposed algorithm
has been simulated in an environment free of obsta-
cles and threats. Besides, Pan etal. (2020) proposed
a modified CIPDE (MCIPDE) with modified JADE
(MJADE) called CIJADE for UAV path planning in a
static 3D environment with obstacles, and the simula-
tion results show that the proposed algorithm, com-
pared with PSO, DE, ABC, Jade, and CIPDE, is more
efficient at determining the ideal or near-optimal fly-
ing path for a UCAV.
Another new hybrid meta-heuristic algorithm pro-
posed by Qu et al. (2020b) is known as reinforce-
ment learning-based grey wolf optimizer algorithm
(RLGWO). The new algorithm is based on GOW
and compared to GOW, SOS, and SA. The results
show that the new algorithm can obtain an effective
and safe path successfully, and it is superior in solv-
ing path planning problems compared to other algo-
rithms, but the search effect is still poor. Finally, Qu
et al. (2020a) proposed a hybrid algorithm called
HSGWO-MSOS that are based on a simplified grey
wolf optimizer (SGWO) and modified symbiotic
organisms search (MSOS) in a static 3D environ-
ment. The results show that the HSGWO-MSOS
method can successfully acquire an effective and safe
route and that it outperforms GWO, SOS, and SA in
addressing the UAV path planning problem.
Conclusion
The systematic review presented in the paper provides
the state of the art of scientific literature about the path
planning optimization of unmanned aerial vehicles
using meta-heuristic algorithms. This paper presents
a review of meta-heuristic algorithms in UAV path
planning based on 30 studies that have been selected
among 340 different studies—the dependences on
inclusion and exclusion criteria and based on the main
question and its sub-questions. Data extraction from
the selected study was collected and available online.
The extracted data contain all the required information
about each study, including article name, authors, pub-
lishers, and more detailed information. The analysis of
the information obtained from each study allowed the
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 26 of 28
Vol:. (1234567890)
finding of progress and gaps in this research area. In
particular, this systematic review has found many gaps
in the area of path planning optimization. There is no
specific algorithm dealing with the dynamic obstacles
and environment; all algorithms have good perfor-
mance and near-optimal solutions in terrain and com-
plex environments. Also, the time for finding the path
increases with the increase in population size and iter-
ation number. Not considering local conditions such
as wind and rain, flying objects are other essential
issues the UAV will face in real-time environments.
In general, most studies use simulation to create maps
(global path planning) and not GPS or local path plan-
ning to create ready maps or sensors while the UAV is
flying. In conclusion, although path planning in UAVs
has improved recently, the gaps mentioned previously
still exist, and the use of UAV has increased in most
fields, especially in military and civil applications.
Author contribution All the authors contributed equally in
preparation of this article. All the authors have seen and agree
with the contents of the article.
Declarations
Conflict of interest The authors declare no competing inter-
ests.
References
Adhikari, D., Kim, E., & Reza, H. (2017). A fuzzy adaptive
differential evolution for multi-objective 3D UAV path
optimization. 2017 IEEE Congress on Evolutionary
Computation, CEC 2017 ‑ Proceedings (pp. 2258–2265).
https:// doi. org/ 10. 1109/ CEC. 2017. 79695 78
Aliyu, M. B. (2017). American journal of engineering research
(AJER) Efficiency of Boolean search strings for informa-
tion retrieval. American Journal of Engineering Research,
6(11), 216–222.
Bäck, T., & Schwefel, H.-P. (1993). An overview of evolutionary
algorithms for parameter optimization. Evolutionary Com‑
putation, 1(1), 1–23. https:// doi. org/ 10. 1162/ evco. 1993.1. 1.1
Beyer, H. G., Schwefel, H. P., & Wegener, I. (2002). How
to analyse evolutionary algorithms. Theoretical Com‑
puter Science, 287(1), 101–130. https:// doi. org/ 10. 1016/
S0304- 3975(02) 00137-8
Biundini, I. Z., Pinto, M. F., Melo, A. G., Marcato, A. L. M.,
Honório, L. M., & Aguiar, M. J. R. (2021). A frame-
work for coverage path planning optimization based on
point cloud for structural inspection. Sensors (switzer‑
land), 21(2), 1–20. https:// doi. org/ 10. 3390/ s2102 0570
Blom, J. D. (2006). Unmanned aerial systems: A historical
perspective, vol.45. Kansas: Combat Studies Institute
Press.
Blum, C., Puchinger, J., Raidl, G., Roli, A., Blum, C., Puchinger,
J., Raidl, G., & Roli, A. (2015). Hybrid metaheuristics in
combinatorial optimization: A survey. Applied Soft Com‑
puting, 11(6), 4135–4151. https:// doi. org/ 10. 1016/j. asoc.
2011. 02. 032
Da Silva Arantes, M., Da Silva Arantes, J., Toledo, C. F. M., &
Williams, B. C. (2016). Ahybrid multi-population genetic
algorithm for UAV path planning. Proceedings of the
2016 Genetic and Evolutionary Computation Conference
(pp. 853–860). https:// doi. org/ 10. 1145/ 29088 12. 29089 19
Danancier, K., Ruvio, D., Sung, I., & Nielsen, P. (2019). Compari-
son of path planning algorithms for an unmanned aerial vehi-
cle deployment under threats. IFAC‑PapersOnLine, 52(13),
1978–1983. https:// doi. org/ 10. 1016/j. ifacol. 2019. 11. 493
de Sales Guerra Tsuzuki, M., de Castro Martins, T., & Takase,
F. K. (2006). Robot path planning using simulated anneal-
ing. IFAC Proceedings Volumes, 39(3), 175–180.https://
doi. org/ 10. 3182/ 20060 517-3- fr- 2903. 00105
Denyer, D., & Tranfield, D. (2009). Producing a system-
atic review. In The SAGE Handbook of Organizational
Research Methods (pp. 671–689).
Dewangan, R. K., Shukla, A., & Godfrey, W. W. (2019). Three
dimensional path planning using Grey wolf optimizer for
UAVs. Applied Intelligence, 49(6), 2201–2217. https://
doi. org/ 10. 1007/ s10489- 018- 1384-y
Du, Y. C., Zhang, M. X., Ling, H. F., & Zheng, Y. J. (2019).
Evolutionary planning of multi-UAV search for missing
tourists. IEEE Access, 7, 73480–73492. https:// doi. org/ 10.
1109/ ACCESS. 2019. 29206 23
Duan, H., Yu, Y., Zhang, X., & Shao, S. (2010). Three-dimension
path planning for UCAV using hybrid meta-heuristic ACO-DE
algorithm. Simulation Modelling Practice and Theory, 18(8),
1104–1115. https:// doi. org/ 10. 1016/j. simpat. 2009. 10. 006
El-henawy, I., & Abdelmegeed, N. A. (2018). Meta-heuristics
algorithms : A survey. International Journal of Computer
Applications,179(22), 45–54.
Fan, M., & Akhter, Y. (2021). A time-varying adaptive inertia
weight based modified PSO algorithm for UAV path plan-
ning. ICREST 2021 ‑ 2nd International Conference on
Robotics, Electrical and Signal Processing Techniques,
May, 573–576. https:// doi. org/ 10. 1109/ ICRES T51555. 2021.
93311 01
Fortin, F. A., De Rainville, F. M., Gardner, M. A., Parizeau, M.,
& Gagńe, C. (2012). DEAP: Evolutionary algorithms made
easy. Journal of Machine Learning Research, 13, 2171–2175.
Game, P. (2020). Bio-inspired Optimization: metaheuris-
tic algorithms for optimization. ArXiv, abs/2003.11637.
https:// doi. org/ 10. 48550/ arXiv. 2003. 11637.
Gandomi, A. H., Yang, X. S., Talatahari, S., & Alavi, A. H.
(2013). Metaheuristic algorithms in modeling and opti-
mization. In Metaheuristic Applications in Structures and
Infrastructures. https:// doi. org/ 10. 1016/ B978-0- 12- 398364-
0. 00001-2
Gavrilas, M. (2010). 2010. to power systems. In Proceedings of
the 12th WSEAS international conference on Mathemati‑
cal methods and computational techniques in electrical
engineering MMACTEE’10). World Scientific and Engi‑
neering Academy and Society (WSEAS) (pp. 95–103).
Wisconsin: Stevens Point.
Ge, F., Li, K., Han, Y., Xu, W., & Wang, Y. (2020). Path plan-
ning of UAV for oilfield inspections in a three-dimensional
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
Page 27 of 28 30
Vol.: (0123456789)
dynamic environment with moving obstacles based on an
improved pigeon-inspired optimization algorithm. Applied
Intelligence, 50(9), 2800–2817. https:// doi. org/ 10. 1007/
s10489- 020- 01650-2
Ghambari, S., Idoumghar, L., Jourdan, L., & Lepagnot, J.
(2019). An improved TLBO algorithm for solving UAV
path planning problem. In 2019 IEEE Symposium Series
on Computational Intelligence, SSCI 2019 (pp. 2261–
2268).https:// doi. org/ 10. 1109/ SSCI4 4817. 2019. 90031 60
Giordan, D., Adams, M. S., Aicardi, I., Alicandro, M., Allasia,
P., Baldo, M., De Berardinis, P., Dominici, D., Godone,
D., Hobbs, P., Lechner, V., Niedzielski, T., Piras, M.,
Rotilio, M., Salvini, R., Segor, V., Sotier, B., & Troilo, F.
(2020). The use of unmanned aerial vehicles (UAVs) for
engineering geology applications. Bulletin of Engineering
Geology and the Environment, 79(7), 3437–3481. https://
doi. org/ 10. 1007/ s10064- 020- 01766-2
Sörensen, K., & Glover F. W. (1986) Metaheuristics. Encyclo‑
pedia of Operations Research and Management Science.
https:// doi. org/ 10. 1007/ 978-1- 4419- 1153-7_ 1167
Gong, Q., Zhou, Y., & Luo, Q. (2011). Hybrid artificial glow-
worm swarm optimization algorithm for solving multi-
dimensional knapsack problem. Procedia Engineering,
15, 2880–2884. https:// doi. org/ 10. 1016/j. proeng. 2011. 08.
542
Huo, L., Zhu, J., Wu, G., & Li, Z. (2020). A novel simulated
annealing based strategy for balanced UAV task assign-
ment and path planning. Sensors, 20(17), 4769. https:// doi.
org/ 10. 3390/ s2017 4769
Jarray, R., & Bouallègue, S. (2020). Multi-verse algorithm based
approach for multicriteria path planning of unmanned aerial
vehicles. International Journal of Advanced Computer Sci‑
ence and Applications, 11(11), 324–334. https:// doi. org/ 10.
14569/ IJACSA. 2020. 01111 42
Jun, M., & D’Andrea, R. (2003). Path planning for unmanned aer‑
ial vehicles in uncertain and adversarial environments. Coop‑
erative Systems. https:// doi. org/ 10. 1007/ 978-1- 4757- 3758-5_6
Kaveh, A., & Farhoudi, N. (2011). A unified approach to param-
eter selection in meta-heuristic algorithms for layout opti-
mization. Journal of Constructional Steel Research, 67(10),
1453–1462. https:// doi. org/ 10. 1016/j. jcsr. 2011. 03. 019
Khalilpourazari, S., Naderi, B., & Khalilpourazary S. (2019).
Multi-objective stochastic fractal search: A powerful algo-
rithm for solving complex multi-objective optimization
problems. Soft Computing, 24(4), 3037–3066. https:// doi.
org/ 10. 1007/ s00500- 019- 04080-6
Khamseh, H. B., Pimenta, L. C. A., & Tôrres, L. A. B. (2009).
Autonomous UAV path planning and estimation. IEEE
Robotics & Automation Magazine, 16(2), 1247–1253.
Kitchenham, B., Pearl Brereton, O., Budgen, D., Turner, M.,
Bailey, J., & Linkman, S. (2009). Systematic literature
reviews in software engineering - a systematic literature
review. Information and Software Technology, 51(1),
7–15. https:// doi. org/ 10. 1016/j. infsof. 2008. 09. 009
Kumar, P., Garg, S., Singh, A., Batra, S., Kumar, N., & You, I.
(2018). MVO-based 2-d path planning scheme for provid-
ing quality of service in UAV environment. IEEE Internet
of Things Journal, 5(3), 1698–1707. https:// doi. org/ 10.
1109/ JIOT. 2018. 27962 43
Liu, X., Du, X., Zhang, X., Zhu, Q., & Guizani, M. (2019).
Evolution-algorithm-based unmanned aerial vehicles path
planning in complex environment. Computers and Elec‑
trical Engineering, 80,106493.https:// doi. org/ 10. 1016/j.
compe leceng. 2019. 106493
Ludwig, S. A. (2016). Improved glowworm swarm optimization
algorithm applied to multi-level thresholding. In 2016 IEEE
Congress on Evolutionary Computation, CEC 2016 (pp.
1533–1540). https:// doi. org/ 10. 1109/ CEC. 2016. 77439 71
Lukač, Z. (2011). Metaheuristic optimization. In Proceedings of
the 11th International Symposium on Operational Research
in Slovenia, SOR 2011, May, 17–22. https:// doi. org/ 10.
4249/ schol arped ia. 11472
Malik, K., & Tayal, A. (2014). Comparison of nature inspired
metaheuristic algorithms. International Journal of Elec‑
tronic and Electrical Engineering, 7(8), 799–802.
Marqués, P. (2016). Advanced UAV aerodynamics, flight sta-
bility and control: An introduction. Advanced UAV Aero‑
dynamics, Flight Stability and Control.https:// doi. org/ 10.
1002/ 97811 18928 691. ch1
Mirjalili, S., & Lewis, A. (2016). The whale optimization
algorithm. Advances in Engineering Software, 95, 51–67.
https:// doi. org/ 10. 1016/j. adven gsoft. 2016. 01. 008
Mirjalili, S., Mirjalili, S. M., & Lewis, A. (2014). Grey wolf
optimizer. Advances in Engineering Software, 69, 46–61.
https:// doi. org/ 10. 1016/j. adven gsoft. 2013. 12. 007
Mirjalili, S., Mirjalili, S. M., & Hatamlou, A. (2016). Multi-
verse optimizer: A nature-inspired algorithm for global
optimization. Neural Computing and Applications, 27(2),
495–513. https:// doi. org/ 10. 1007/ s00521- 015- 1870-7
Niu, P., Niu, S., Liu, N., & Chang, L. (2019). The defect of
the grey wolf optimization algorithm and its verification
method. Knowledge‑Based Systems, 171, 37–43. https://
doi. org/ 10. 1016/j. knosys. 2019. 01. 018
Oz, I., Topcuoglu, H. R., & Ermis, M. (2013). A meta-heuristic
based three-dimensional path planning environment for
unmanned aerial vehicles. SIMULATION, 89(8), 903–
920. https:// doi. org/ 10. 1177/ 00375 49712 456419
Pan, J. S., Liu, N., & Chu, S. C. (2020). A hybrid differential
evolution algorithm and its application in unmanned com-
bat aerial vehicle path planning. IEEE Access, 8, 17691–
17712. https:// doi. org/ 10. 1109/ ACCESS. 2020. 29681 19
Pandey, P., Shukla, A., & Tiwari, R. (2018). Three-dimensional
path planning for unmanned aerial vehicles using glow-
worm swarm optimization algorithm. International Jour‑
nal of Systems Assurance Engineering and Management,
9(4), 836–852. https:// doi. org/ 10. 1007/ s13198- 017- 0663-z
Qu, C., Gai, W., Zhong, M., & Zhang, J. (2020b). A novel
reinforcement learning based grey wolf optimizer algo-
rithm for unmanned aerial vehicles (UAVs) path planning.
Applied Soft Computing Journal, 89, 106099. https:// doi.
org/ 10. 1016/j. asoc. 2020. 106099
Qu, C., Gai, W., Zhang, J., & Zhong, M. (2020a). A novel
hybrid grey wolf optimizer algorithm for unmanned aerial
vehicle (UAV) path planning. Knowledge‑Based Systems,
194(xxxx), 105530. https:// doi. org/ 10. 1016/j. knosys. 2020a.
105530
Ri, R., Sdwk, R., Dojrulwkpv, S., Wkh, R. I., Nqrzq, Z.,
Edvlf, D. Q. G., Dojrulwkp, K., & Lv, A. (n.d.). &Rps‑
dulvrq Ri Rswlpdo Sdwk Sodqqlqj Dojrulwkpv. 6, 6–9.
Roberge, V., Tarbouchi, M., & Labonte, G. (2018). Fast
genetic algorithm path planner for fixed-wing military
UAV using GPU. IEEE Transactions on Aerospace and
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

Environ Monit Assess (2023) 195:30
1 3
30 Page 28 of 28
Vol:. (1234567890)
Electronic Systems, 54(5), 2105–2117. https:// doi. org/
10. 1109/ TAES. 2018. 28075 58
Rudas, I. J., & Tar, J. K. (2010). Computational intelligence
for problem solving in engineering. IECON Proceedings
(industrial Electronics Conference), 1317–1322. https://
doi. org/ 10. 1109/ IECON. 2010. 56754 91
Said, G. A. E -N., Mahmoud, A. M., & El-Horbaty, E.-S. M.
(2014). A comparative study of meta-heuristic algorithms for
solving quadratic assignment problem. International Journal
of Advanced Computer Science and Applications, 5(1), 1–6.
https:// doi. org/ 10. 14569/ ijacsa. 2014. 050101
Saka, M. P., Doǧan, E., & Aydogdu, I. (2013). Analysis of swarm
intelligence-based algorithms for constrained optimization.
Swarm Intelligence and Bio‑Inspired Computation, 25–48.
https:// doi. org/ 10. 1016/ B978-0- 12- 405163- 8. 00002-8
Shao, S., Peng, Y., He, C., & Du, Y. (2020). Efficient path planning
for UAV formation via comprehensively improved particle
swarm optimization. ISA Transactions, 97(xxxx), 415–430.
https:// doi. org/ 10. 1016/j. isatra. 2019. 08. 018
Shin, J. J., & Bang, H. (2020). UAV path planning under
dynamic threats using an improved PSO algorithm. Inter‑
national Journal of Aerospace Engineering, 2020. https://
doi. org/ 10. 1155/ 2020/ 88202 84
Sudhakar, S., Vijayakumar, V., Sathiya Kumar, C., Priya, V.,
Ravi, L., & Subramaniyaswamy, V. (2020). Unmanned
aerial vehicle (UAV) based forest fire detection and moni-
toring for reducing false alarms in forest-fires. Computer
Communications, 149, 1–16. https:// doi. org/ 10. 1016/j.
comcom. 2019. 10. 007
Tang, A. D., Han, T., Zhou, H., & Xie, L. (2021). An improved
equilibrium optimizer with application in unmanned aerial
vehicle path planning. Sensors, 21(5), 1–21. https:// doi.
org/ 10. 3390/ s2105 1814
Hongtao Tao, Zheng Wang, J. L. (2014). Three-dimensional
path planning for unmanned aerial vehicles based on
multi-objective genetic algorithm. Proceedings of the
33rd Chinese Control Conference, 8617–8621.
Tsourdos, A., White, B., & Shanmugavel, M. (2010). Coop-
erative path planning of unmanned aerial vehicles. In
Cooperative Path Planning of Unmanned Aerial Vehicles.
https:// doi. org/ 10. 1002/ 97804 70974 636
Turker, T., Sahingoz, O. K., & Yilmaz, G. (2015). 2D path plan-
ning for UAVs in radar threatening environment using
simulated annealing algorithm. In 2015 International Con‑
ference on Unmanned Aircraft Systems, ICUAS 2015 (pp.
56–61). https:// doi. org/ 10. 1109/ ICUAS. 2015. 71522 75
Valavanis, K. P., & Vachtsevanos, G. J. (2015). Handbook of
unmanned aerial vehicles. vehicles, vol. 1. Dordrecht:
Springer . https:// doi. org/ 10. 1007/ 978- 90- 481- 9707-1
Valavanis, K. P. (2017). Unmanned Aircraft Systems Challenges
in Design for Autonomy. In 11th International Workshop
on Robot Motion and Control (RoMoCo(pp.73–86).https://
doi. org/ 10. 1109/ RoMoCo. 2017. 80038 96
Wang, G., Guo, L., Duan, H., Wang, H., Liu, L., & Shao, M.
(2012). A hybrid metaheuristic DE/CS algorithm for
UCAV three-dimension path planning. The Scientific
World Journal, 2012. https:// doi. org/ 10. 1100/ 2012/ 583973
Wang, C., Liu, P., Zhang, T., & Sun, J. (2018). The adaptive vor-
tex search algorithm of optimal path planning for forest fire
rescue UAV. In Proceedings of 2018 IEEE 3rd Advanced
Information Technology, Electronic and Automation Control
Conference, IAEAC 2018, Iaeac (pp 400–403). https:// doi.
org/ 10. 1109/ IAEAC. 2018. 85777 33
Wohlin, C. (2014). Guidelines for snowballing in systematic
literature studies and a replication in software engineer-
ing. ACM International Conference Proceeding Series.
https:// doi. org/ 10. 1145/ 26012 48260 1268
Xin, J., Zhong, J., Yang, F., Cui, Y., & Sheng, J. (2019).
An improved genetic algorithm for path-planning of
unmanned surface vehicle. Sensors (switzerland), 19(11),
1–23. https:// doi. org/ 10. 3390/ s1911 2640
Yang, L., Qi, J., Song, D., Xiao, J., Han, J., & Xia, Y. (2016).
Survey of robot 3D path planning algorithms. Journal of
Control Science and Engineering, 2016,1–22. https:// doi.
org/ 10. 1155/ 2016/ 74269 13
Yang, Liuqing, Guo, J., & Liu, Y. (2020). Three-dimensional
UAV cooperative path planning based on the MP-CGWO
algorithm. International Journal of Innovative Comput‑
ing, Information and Control, 16(3), 991–1006. https://
doi. org/ 10. 24507/ ijicic. 16. 03. 991
Yu, X., Li, C., & Yen, G. G. (2021). A knee-guided differential
evolution algorithm for unmanned aerial vehicle path plan-
ning in disaster management. Applied Soft Computing, 98,
106857. https:// doi. org/ 10. 1016/j. asoc. 2020. 106857
Yuan, C., Liu, Z., & Zhang, Y. (2017). Fire detection using
infrared images for UAV-based forest fire surveillance. In
2017 International Conference on Unmanned Aircraft Sys‑
tems, ICUAS 2017 (pp. 567–572). https:// doi. org/ 10. 1109/
ICUAS. 2017. 79913 06
Zekui, Q., Rui, W., Xiwang, D., Qingdong, L., Dongyang, F.,
& Zhang, R. (2018). Three-dimensional path planning for
unmanned aerial vehicles based on the developed RRT
algorithm. 2018 IEEE CSAA Guidance. Navigation and
Control Conference, CGNCC, 2018, 8617–8621. https://
doi. org/ 10. 1109/ GNCC4 2960. 2018. 90189 65
Zelinka, I. (2015). A survey on evolutionary algorithms
dynamics and its complexity - Mutual relations, past, pre-
sent and future. Swarm and Evolutionary Computation,
25, 2–14. https:// doi. org/ 10. 1016/j. swevo. 2015. 06. 002
Zhang, S., Zhou, Y., Li, Z., & Pan, W. (2016). Grey wolf opti-
mizer for unmanned combat aerial vehicle path planning.
Advances in Engineering Software, 99, 121–136. https://
doi. org/ 10. 1016/j. adven gsoft. 2016. 05. 015
Zheng, X., Bao, C., & He, Z. (2018). Design of simulation test
platform for UAV flight control system. Journal of Phys‑
ics: Conference Series, 1069(1), 012022. https:// doi. org/
10. 1088/ 1742- 6596/ 1069/1/ 012022
Zhou, X., Gao, F., Fang, X., & Lan, Z. (2021). Improved bat
algorithm for UAV path planning in three-dimensional
space. IEEE Access, 9, 20100–20116. https:// doi. org/ 10.
1109/ ACCESS. 2021. 30541 79
Publisher’s note Springer Nature remains neutral with regard
to jurisdictional claims in published maps and institutional
affiliations.
Springer Nature or its licensor (e.g. a society or other partner)
holds exclusive rights to this article under a publishing agreement
with the author(s) or other rightsholder(s); author self-archiving of
the accepted manuscript version of this article is solely governed
by the terms of such publishing agreement and applicable law.
Content courtesy of Springer Nature, terms of use apply. Rights reserved.

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”),
for small-scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are
maintained. By accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use
(“Terms”). For these purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or
a personal subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or
a personal subscription (to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the
Creative Commons license used will apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data
internally within ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking,
analysis and reporting. We will not otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of
companies unless we have your permission as detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that
Users may not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to
circumvent access control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil
liability, or is otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by
Springer Nature in writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer
Nature journal content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates
revenue, royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain.
Springer Nature journal content cannot be used for inter-library loans and librarians may not upload Springer Nature journal
content on a large scale into their, or any other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any
information or content on this website and may remove it or features or functionality at our sole discretion, at any time with or
without notice. Springer Nature may revoke this licence to you at any time and remove access to any copies of the Springer Nature
journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express
or implied with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or
warranties imposed by law, including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be
licensed from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other
manner not expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com