Highlights in Science, Engineering and Technology EMIS 2024

Volume 119 (2024)

Research progress of path planning algorithm for mobile robot

navigation
Xueyi Zhang *
Department of Shanghai Maritime University, Shanghai, China
* Corresponding author: [email protected]
Abstract. The main research contents of mobile robot technology include: navigation and positioning,
path planning and motion control. Path planning is an important branch of mobile robot research. Its
task is to find an optimal path in its workspace according to some optimization criteria. Many different
path planning algorithms have been developed to solve the corresponding problems, such as
improving the safety of the path, enhancing the working ability of the robot and improving the
efficiency of path planning. However, a single original path planning algorithm often has different
advantages and disadvantages. Therefore, many scholars have made improvements for the defects
of various algorithms According to the design principle of path planning algorithm, this article divides
the mainstream path planning algorithm into three types: graph-search-based algorithm, potential
field-based algorithm and intelligent bionic-based algorithm. The principle of the path planning
algorithm mentioned in this paper is briefly described and the characteristics of different improved
algorithms are summarized. The future research directions are prospected from three aspects:
adaptability and intelligence, fusion of multiple optimization algorithms and interdisciplinary
technologies integration. This article provides a reference for relevant researchers to understand the
ideas of related improved algorithms.
Keywords: mobile robot, path planning, graph-search-based algorithm, intelligent bionic-based
algorithm, potential field-based algorithm.

1. Introduction
With the continuous updating of science and technology, mobile robots have not only been the key
research topic in the field of artificial intelligence, but have also penetrated into various fields of life.
The main research of mobile robot technology includes three parts: navigation and localization, path
planning and motion control [1]. Path planning as a key technology in the field of robotics has also
been a hot area of research. The optimal path planning for mobile robots aims to intelligently plan the
best travel path from the starting point to the target point in their complex and variable working
environment. It is always based on certain optimization criteria, such as minimizing the cost of the
work, shortest path length or shortest travel time [2]. Therefore, path planning for mobile robots is
critical. For example, mobile robots are responsible for the transportation and handling of materials
in automated factories. Path planning can ensure that robots travel quickly and safely in complex
workshop environments, reducing material transportation time and increasing production efficiency.
If the path planning is improper, the robot may collide with equipment or personnel, resulting in
production interruptions and safety hazards. Among the path planning algorithms, A* Algorithm,
Dijkstra's Algorithm, Ant Colony Algorithm and Exploring Random Tree Algorithm are commonly
used algorithms for path planning [3-6]. According to different classification criteria, path planning
algorithms can be categorized into global path planning algorithms and local path planning algorithms
[7]. Another, path planning algorithms can also be categorized into graph-based search, potential
field-based, and intelligent bionic path planning algorithm [8]. The proposal of these algorithms has
greatly promoted technological innovation and industrial upgrading in the field of path planning.
Compared with the more primitive algorithms, the above algorithms can improve the efficiency and
accuracy of path planning to different degrees, optimize the use of resources and reduce the cost.
Considering the safety of the actual work, they can also reduce the risk of accidents.
Through optimizing different algorithms, the obstacle avoidance ability, search efficiency and
planning accuracy of mobile robots can be improved. However, with the deepening study of various
366
Highlights in Science, Engineering and Technology EMIS 2024
Volume 119 (2024)

algorithms, different algorithms reflect their superiority and limitations. For example, the traditional
RRT algorithm has low running efficiency, the planned path is not optimal due to the large
randomness and the artificial potential field method is prone to fail to find the path due to low gravity,
etc. [9-10]. Therefore, in practical applications, different algorithms are influenced by various factors,
reflecting different characteristics.
In this article, the path planning algorithms for mobile robots are categorized into graph-based
search, potential field-based, and intelligent bionic path planning algorithm according to their
characteristics [1,8]. The different path planning optimization algorithms in recent years are
introduced, while their characteristics are summarized, and finally outlook for future path planning
algorithm optimization.

2. Main Body
2.1. Path Planning Algorithm Based on Graph Search
Graph search algorithms are algorithms for traversing or searching graph structures. In path
planning, graph-based search algorithms are usually used in low dimensional spaces, typically in the
form of raster maps to completely represent the entire environment map. They have the characteristics
of simple, effective, and easy to modify [11].
2.1.1. A* Algorithm
The A* algorithm was published by Hart et al in 1968 [3]. It is a heuristic search algorithm that
reduces unnecessary searches and improves the search efficiency. It is mainly used in graph search
to find the least costly path from the starting point to the target point.
Compared to the traditional A* algorithm, Yang et al. proposed three improvements: excluding all
forced neighboring nodes, extracting the crucial turning points to reducing path redundancies and the
number of robot's steering, and using Bezier curves to achieve segmental smoothing [12]. Four sets
of simulations from different perspectives show that the above improvements compared with the
original A* algorithm, the security exhibits an average 33.68% enhancement in path safety on average
and the number of turning points is reduced by 37.00% on average. Additionally, the path is smoother.
Wang Hai-qun et al. proposed a new dynamic measurement heuristic A * algorithm [13]. He
established a grid map and added new adjustment factors to address the situation where there are
many inflection points in the path obtained by the traditional A * algorithm. After simulation
experiment data analysis, the number of inflection points has been reduced. In addition, it can
significantly improve search efficiency and shorten path length. To solve problems about large search
range and collision path in global path planning, Zhang Wei et al. proposed an improved A*algorithm
is devised ground on thought of adaptive search distance [14]. The search distance is adaptively
adjusted according to the complexity of the environment to reduce the number of extended search
nodes and search time. Build up an anti-collision distance function to the cost function of the
improved A * algorithm. Through simulation and actual testing, the results show that the improved
A * algorithm optimizes the number of search nodes, iteration times, and total turning angle in the
path planning stage. Xu Jianmin et al. proposed a fusion algorithm based on the A* algorithm of the
Multi-scale Map Approach and the improved Dynamic Window Approach (DWA) algorithm [15].
The fusion algorithm can address the issues of sharply increasing computational and time costs and
poor flexibility in large-scale complex environments. The A * algorithm is developed by establishing
a multi-scale set, adding an obstacle proportion factor to the heuristic function and using the Floyd
algorithm to optimize the nodes. Besides, the heading angle adaptive adjustment strategy and parking
wait state are added to optimize the dynamic window method. After the ROS simulation and actual
vehicle experiments, the fusion algorithm enables mobile robot to have higher flexibility and
adaptability, and it can greatly shorten search time and improve search efficiency.

367
Highlights in Science, Engineering and Technology EMIS 2024
Volume 119 (2024)

2.1.2. Dynamic A* Algorithm

The D*(Dynamic A*) algorithm was proposed by Sven Koenig and Maxim Likhachev in 2002
[16]. It is suitable for dynamic environments and continuously updates the proxy value of each node
in the path. When there are new node changes, it can immediately determine the new shortest path
[17].
In unknown and complex environment, traditional D* algorithm have problems of blind search,
low computational efficiency, and difficulty in obtaining safe trajectories. Ju Muhan et al. proposed
an improved D* algorithm, incorporating the idea of jump point search into front-end path planning
to improve planning efficiency [18]. Based on the B-spline, the back-end construct a quadratic
planning problem to simplify the difficulty of solving. The results show that comparing with the
traditional D * algorithm, Focused D * algorithm, and directed D * algorithm, the number of search
nodes and search time have been optimized to varying degrees. In order to solve the problems of low
algorithm efficiency and complex inflection points in the use of traditional D* algorithm for the
unmanned aerial vehicle, Wang Xiaoshuai et al. adopted a variable step size approach in selecting
extension nodes which accelerates the traversal speed of nodes [19]. In addition, the fusion formula
of Chebyshev distance and Manhattan distance replaces the traditional Euclidean distance, reducing
computational complexity and improving efficiency. Through MATLAB simulation, the
performance of the improved D* algorithm in three-dimensional experimental environments is more
efficient, path safety is improved, and path complexity is reduced.
2.2. Path planning algorithm based on potential field
The Artificial Potential Field (APF) method was proposed by Khatib in 1985. This algorithm sets
the target point to have attraction on the robot, while obstacles have repulsion on the robot [20]. The
direction of the resultant force represents the expected direction of motion, and the magnitude of the
resultant force represents the desired speed. This algorithm has the characteristics of simple structure
and convenient real-time control [21].
To solve the local minimization and trajectory oscillation of mining mobile robots in complex
environments, Zhang Desheng proposed an improved APF method [22]. To avoid falling into the
local minimum position, this method introduces virtual target points to guide the robot escape from
the dead zone. By using a fuzzy controller to determine the potential field situation the method can
enhance the trajectory smoothness of robot. Through MATLAB simulation, this improvement
method has higher environmental adaptability and smoother path. Wu Meihua introduces the relative
velocity and acceleration factors between the mobile robot and the target point into the gravity
function [23]. This ensures that the mobile robot maintains the same motion trend with the target
point. Introducing the radius of the obstacle and the radius of the robot to avoid collisions between
mobile robots and obstacles. In order to avoid the robot getting stuck in local minimization, the center
line method is proposed to change the resultant force on the robot. The improved algorithm has been
proven effective through MATLAB simulation. Aiming at the difficulty of path planning for mobile
robots in complex environments, Shi Weiguo et al. proposed a mixed ant colony potential field
algorithm (MACPF) [24]. Firstly, multi-factor heuristic function and new ant travel mechanism are
used to improve path quality. Secondly, the adaptive volatilization coefficient and dynamic weight
coefficient are designed to accelerate convergence speed. Thirdly, traditional artificial potential field
methods are prone to getting stuck in local minimization, unreachable target and excessive obstacle
avoidance. This method solves these problems by introducing the concepts of virtual target point,
relative distance and safe distance. Through simulation analysis, the MACPF algorithm has
demonstrated strong environmental adaptability and path planning efficiency. NíJianyun et al.
proposed an improved APF method to address the issues of unreachable targets and local
minimization [25]. By introducing the distance factor between the robot and the target point, an
elliptical target strategy and an elliptical tracking strategy, the optimization goal is achieved. In
addition, a cubic uniform B spline curve is introduced to solve the problems of low turning efficiency

368
Highlights in Science, Engineering and Technology EMIS 2024
Volume 119 (2024)

and unbalanced turning points. The feasibility and effectiveness of this improved method have been
confirmed through simulation.
2.3. Path Planning Algorithm Based on Intelligent Bionics
Path planning algorithm based on intelligent bionics are algorithms that mimic certain group
behaviors or biological structures and utilize swarm intelligence. Its main characteristics include
biomimetic features, simplicity and emergence, robustness and self-organization. [26].
2.3.1. Ant Colony Optimization
Ant Colony Optimization (ACO) was proposed by Italian scholar Dorigol in 1992 [27]. It
simulates the path optimization method of ants searching for food in nature and uses pheromone
precipitation and pheromone evaporation rules to search for the best path.
Xu Jianmin et al. proposed a multi-level field of view adaptive ant colony optimization
(MLFVAACO) [28]. Firstly, it extends the two-level field of view on the basis of traditional AOC to
make the planned path smoother. Secondly, an adaptive global initial pheromone update strategy is
proposed. Thirdly, the deadlock ants in the algorithm iteration process are optimized to increase the
diversity of search solutions. Finally, the state transition rule of ants is employed to avoid ants from
falling into the local optimal solution. The simulation results show that the MLFVAACO algorithm
performs well in terms of optimal path, ant utilization and search efficiency, while it has high
adaptability to complex environments. Aiming at the slow convergence speed and tendency to fall
into the local optimal solution, Xue Xiang et al. proposed a new improved algorithm [29]. Firstly, a
trend heuristic function was established and the Cauchy distribution function was introduced on the
basis of this function. Then, improve the distance heuristic function to make path selection more
purposeful. Thirdly, Improve the pheromone volatilization factor to accelerate convergence speed.
Through MATLAB simulation, this improved algorithm shows significant improvements in
minimum path length, convergence speed, and smoothness compared to traditional ACO algorithms.
To improve the path planning performance of rescue robots in the fire scene, Ding Jinhu proposed
several measures [30]. Build indoor environment models by using BIM technology. Optimize the
process of extracting and updating pheromones. Improve pheromone volatilization factor. Add
probability factors to avoid obstacles. Set specific conditions to make ants jump out of the loop.
Finally, determine the suitability of different escape speeds and routes for trapped individuals and
select appropriate escape points. Verified by simulation experiments, this improved algorithm has
fewer iterations, stronger convergence, and is suitable for complex environments.
2.3.2. Genetic Algorithms
Genetic algorithms (GA) were proposed by John Holland in the early 1970s [31]. This algorithm
imitates the evolution and genetic mechanisms of nature and use operations such as screening,
hybridization, and mutation. It gradually generates and optimizes the next generation of potential or
optimal solutions and ultimately achieve a gradual optimization of the effect [32].
WU Yue'an et al. optimized the order of tilt-rotor unmanned aerial vehicle (TRUAV) coverage
paths through the minimum spanning and back-and-forth path generation algorithms [33]. Adopting
a fishtail-shaped obstacle avoidance strategy to avoid obstacles within the area. Heuristic search for
coverage path order is performed using the Dubins-based enhanced genetic algorithm. Improve the
generation of the initial population, design of the fitness function, crossover operator, and compilation
operator. According to the simulation results, the area coverage efficiency of the tiltrotor unmanned
aerial vehicle is significantly improved, and the length of the coverage path is greatly reduced. When
using traditional genetic algorithms for path planning of goods transportation in logistics transit
warehouses, mobile robots have problems with premature convergence and poor optimization ability.
Liu Zhi-hai et al. improved it through three genetic operators: selection, crossover, and mutation [34].
The selection adopts a probability based on roulette wheel method. Adopting a heuristic crossover
operator to ensure that the crossover offspring are likely to evolve towards the optimal solution. A
mutation operator with adaptive strategy adjustment is proposed, which uses inverse trigonometric

369
Highlights in Science, Engineering and Technology EMIS 2024
Volume 119 (2024)

functions to appropriately adjust the mutation probability. Based on MATLAB simulation analysis,
the number of iterations, planning time, and path time have all been significantly optimized.
2.3.3. Particle Swarm Optimization
Particle Swarm Optimization (PSO) was proposed by Eberhart et al. in 1995 [35]. Its core idea is
that each individual in a group continuously changes their search pattern based on their own
experience and the experience of other individuals. Each individual (particle) only has two attributes:
position and velocity. The particle swarm optimization has strong robustness [36].
Wu Nini et al. proposed a path planning method based on autonomous learning particle swarm
algorithm to reduce the path length of mobile robots [37]. Establish a path planning model with the
shortest path as the optimization objective. Use obstacle expansion method to prevent collision of
robots. A learning strategy pool composed of multiple particle-learning strategies is introduced,
enabling particles to have autonomous learning ability. Through the algorithm performance test, the
improved algorithm has stronger optimization ability and greater stability. Zhao Di et al. proposed an
exchanging PSO (EXPSO) algorithm based on global and local information exchange during search
[38]. A new exchange item is added for the local information and global information in the particle
search process. Use the penalty function to avoid obstacles. Smooth the path using cubic spline
function. The simulation results show that the speed and convergence precision of the EXPSO
algorithm in searching for the global optimal solution have been optimized. Hao Kun et al. proposed
a path planning method based on region search-adaptive particle swarm optimization algorithm (RS-
APSO) [39]. The regional search algorithm is used to solve the problem of long exploration time and
reduce invalid information in the map. Two variable operators are proposed to adjust the inertia
weight factor and improve the acceleration factor adaptively. Finally, a new dynamic obstacle
avoidance strategy is proposed to ensure the safety of robots. The simulation results show that the
RS-APSO algorithm has stronger optimization capabilities in static environments and can
successfully generate safe path in dynamic environments.

3. Conclusion
This article briefly introduces the basic principles of each path planning algorithm involved, and
summarizes the research progress of relatively new improved algorithms. The selection of literature
is based on different types of mobile robots, such as AGV, mining mobile robot, UAV, etc. The
improved algorithms in the literature put forward different improvement measures for the defects of
the original algorithm. The simulation and experiments results show that compared with the original
algorithm, these improved algorithms have different degrees of improvement and optimization. From
the perspective of current development status and practical needs, research on path planning
algorithms for mobile robot in the future should focus more on the following aspects.
Most existing research is based on simulation environments using software such as MATLAB,
Webots and ROS. However, the complexity of practical environments is often determined by different
factors, such as spatial irregularity and disorder. The actual dynamic factors such as lighting, weather,
temperature, and other environmental changes can also affect the final effectiveness of the algorithm.
When encountering unexpected scenarios, the accuracy and security of the path cannot be guaranteed.
Therefore, future research on improving algorithms should focus more on enhancing their adaptability
and intelligence. To obtain the optimal path by making autonomous adjustments to complex and
variable environments.
Different algorithms have their own advantages and disadvantages. A single path planning
algorithm cannot meet the complexity of current practical application scenarios. People often require
path planning algorithms that meet a combination of high efficiency, safety, and low consumption
requirements. Introducing corresponding optimization algorithms at different stages of path planning
can solve corresponding problems. This can effectively improve the reliability and practicality of path
planning.

370
Highlights in Science, Engineering and Technology EMIS 2024
Volume 119 (2024)

Most existing improvements are modifications to the algorithm function itself. In addition, it can
cooperate with technologies of other disciplines. For example, multi-sensor information and data
fusion can provide more detailed data of mobile robots. Combined with control strategy and stability
analysis technology in control engineering, the efficiency and performance of path planning algorithm
can be improved. Through the coordinated control strategy, the cooperative operation of obstacle
avoidance and path optimization among multiple robots can be realized. To sum up, technology
integration with other disciplines can further improve the practical feasibility of the path planning.

References
[1] Cui wei, Zhu Fazheng. Review of path planning algorithms for robot navigation [j]. computer engineering
and applications, 2023, 59 (19): 10 - 20.
[2] WANG K, XU J, SONG K, et al. Informed anytime Bidirectional fast marching tree for optimal motion
planning in complex cluttered environments [J]. Expert Systems with Applications, 2023, 215: 119263.
[3] HART P E, NILSSON N J, RAPHAEL B.A formal basis for the heuristic determination of minimum cost
paths [J]. IEEE Transactions on Systems Science and Cybernetics (T-SSC), 1968, 4 (2): 100 - 107.
[4] DIJKSTRA E W. A note on two problems in connexion with graphs [J]. Numerische Mathematik, 1959,
1 (1): 269 - 271.
[5] DORIGO M, MANIEZZO V, COLORNI A. Positive feedback as a search strategy [R]. Technical Report:
91 - 016, 1999.
[6] Liu Chong, Liu benxue, Lu eucalyptus, et al. Indoor mobile robot path planning based on improved RRT
algorithm [j]. modular machine tools and automated processing technology, 2023 (10): 20 - 23.
[7] Wang hejing, Wang lina. Review of robot path planning algorithms [j]. Journal of Guilin University of
technology, 2023,43 (01): 137 - 147.
[8] Guo xiaoying, Zhao shuping. Review of algorithm research in robot path planning[j/ol]. radio engineering,
1-9 [2024-08-02] http: //kns. cnki. net/kcms/detail/13.1097. TN. 20240527. 1145. 002. html.
[9] Dengyizhao, tuhaiyan, songmingjun. Robot path planning algorithm based on improved RRT [j]. modular
machine tools and automatic processing technology, 2024, (06): 6 - 11.doi: 10.13462/j. cnki. mmtamt.
2024. 06. 002.
[10] Xin Peng. Research on mobile robot path planning based on a * algorithm [d]. Hebei University of
technology, 2022.doi: 10. 27104/d. cnki. ghbjy. 2022. 000186.
[11] Miao Yinjun, Shi Wei. Research on unmanned vehicle path planning based on improved A-star algorithm
[j]. computer knowledge and technology, 2024, 20 (03): 4 - 7. doi: 10.14004/j. cnki. ckt. 2024. 0161.
[12] Yang Guo, Wu Xiao, Xiao Ruqi, et al. Safe and efficient indoor global path planning based on improved
a~ (*) algorithm [j/ol]. Journal of electronic measurement and instrumentation, 1-13 [2024-08-07]
http://kns.cnki.net/kcms/detail/11. 2488. tn. 20240729. 1033. 004. html.
[13] Wanghaiqun, Gan Chengtong, wangfubin. Unmanned vehicle path planning based on improved a *
algorithm [j/ol]. mechanical design and manufacturing, 1-7 [2024-08-09]
https://doi.org/10.19356/j.cnki.1001-3997. 20240725. 026.
[14] Zhang Wei, Zhang Zezhou, Wang Wei. Research on improved a * algorithm for adaptive search distance
[j/ol]. mechanical design and manufacturing, 1-5 [2024-08-09] https://doi.org/10.19356/j.cnki.1001-3997.
20240327. 004.
[15] Xujianmin, song Lei, dengdongdong, et al. Mobile robot path planning based on the fusion of multi-scale
a * and optimized DWA algorithm [j/ol]. Journal of system simulation, 1-12 [August 11, 2024]
https://doi.org/10.16182/j.issn1004731x.joss.223 - 1089.
[16] Koenig S, Likhachev M.Real-time adaptive A[C]//Proceedings of the 5th International Joint Conference
on Autonomous Agents and Multiagent Systems, 2006: 281 - 288.
[17] Zhaoxuejian, Yehao, Jiawei, et al Review of AGV path planning and obstacle avoidance algorithm [j].
small microcomputer system, 2024,45 (03): 529 - 541.doi: 10.20009/j.cnki.21-1106/tp.2023 - 0418.

371
Highlights in Science, Engineering and Technology EMIS 2024
Volume 119 (2024)

[18] Ju Muhan, Liu Wanke, Hu Jie, et al. Robot motion planning in unknown scenes based on improved d~*
algorithm [j]. Journal of navigation and positioning, 2024,12 (03): 145 - 153.doi: 10.16547/j.cnki.10 -
1096. 20240317.
[19] WangXiaoShuai, zhuqixin, zhuyonghong. UAV 3D path planning based on improved d~* algorithm [j].
Journal of Xi'an University of technology, 2023, 37 (03): 83 - 91.doi: 10.13338/j.issn.1674 - 649x. 2023.
03. 012.
[20] KHATIB O. Real-time obstacle avoidance for manipulators and mobile robots [J]. The International
Journal of Robotics Research, 1986, 5 (1): 90 - 98.
[21] Huofengcai, Chi Jin, huangzijian, et al. Overview of mobile robot path planning algorithms [j]. Journal of
Jilin University (Information Science Edition), 2018,36 (06): 639 - 647.doi: 10.19292/j. cnki. jdxxp. 2018.
06. 007.
[22] Zhangdesheng. Path planning of Mining Mobile Robot Based on artificial potential field method and
fuzzy control [j]. coal mine machinery, 2024,45 (06): 199 - 202.doi: 10.13436/j.mkjx.202406057.
[23] Wu Meihua. Research on mobile robot path planning based on improved artificial potential field method
[j]. Heilongjiang science, 2024, 15 (08): 81 - 84.
[24] Shi Weiguo, Ning Ning, song Cunli, et al. Mobile robot path planning based on ant colony algorithm and
artificial potential field method [j]. Journal of agricultural machinery, 2023, 54 (12): 407 - 416.
[25] Nijianyun, duhelei, Gu Haiqing, et al. Research on path planning of mobile robot based on improved
artificial potential field method [j]. Journal of Chongqing University of Technology (NATURAL
SCIENCE), 2023, 37 (11): 247 - 256.
[26] Chen Jun, Shen Qiqi. A review of research on path planning algorithm of AGV [j]. automation and
instrumentation, 2023, (09): 8 - 15.doi: 10.14016/j.cnki.1001 - 9227.2023.09.008.
[27] Colorni A, Dorigo M, Maniezzo V. An investigation of some properties of an ant algorithm [C]//Ppsn,
1992, 92 (1992): 509 - 520.
[28] Xujianmin, dengdongdong, song Lei, et al. Mobile robot path planning based on multi-level vision
adaptive ant colony algorithm [j/ol]. Journal of agricultural machinery, 1-13 [2024-08-16]
http://kns.cnki.net/kcms/detail/11.1964.s. 20240712. 1521. 007. html.
[29] Xuexiang, zhuqixin, zhuyonghong. Robot path planning based on improved ant colony algorithm [j/ol].
Journal of Xi'an University of technology, 1-8 [2024-08-16] http://kns.cnki.net/kcms/detail/61. 1471. N.
20240624. 1042. 002. html.
[30] Dingjinhu. Path planning method of fire rescue robot based on improved ant colony algorithm [j]. science
and technology innovation and application, 2024, 14 (14): 133 - 136.doi: 10.19981/j.cn23-1581/g3. 2024.
14. 032.
[31] Jeffrey R. Sampson. Adaptation in Natural and Artificial Systems (John H. Holland) [J]. Society for
Industrial & Applied Mathematics (siam),1976, 18 (3): 529 - 530.
[32] John H Holland.Adaptation in natural and artificial systems:an introductory analysis with applications to
biology, control, and artificial intelligence [M]. Massachusetts: The MIT Press: 1992 - 04 - 29.
[33] Wuyuean, duchangping, Yang Rui, et al. Area coverage path planning for tilt rotor UAV Based on
improved genetic algorithm [j/ol]. Journal of Zhejiang University (Engineering Edition), 1-10 [2024-08-
16] http://kns.cnki.net/kcms/detail/33. 1245. T. 20240725. 0958. 002. html.
[34] Liuzhihai, Gao long, hanwenyu, et al. Research on mobile robot path planning based on improved genetic
algorithm [j]. manufacturing automation, 2024, 46 (05): 26 - 30.
[35] Eberhart R, Kennedy J.A new optimizer using particle swarm theory[C]//MHS′95, Proceedings of the 6th
International Symposium on Micro Machine and Human Science, IEEE, 1995: 39 - 43.
[36] Lvbinghang. Research on AGV path planning algorithm based on ant colony particle swarm optimization
fusion algorithm [d]. Haikou: Hainan University, 2022.
[37] Wu Nini, Wang xiuxin. Autonomous learning particle swarm optimization method for mobile robot
navigation path [j]. mechanical design and manufacturing, 2024, (07): 342 - 346. doi: 10.19356/j. cnki.
1001 - 3997. 20240322. 006.

372
Highlights in Science, Engineering and Technology EMIS 2024
Volume 119 (2024)

[38] Zhaodi, hekeqin, zhaozugao. Mobile robot path planning based on improved particle swarm optimization
algorithm [j]. sensors and Microsystems, 2023, 42 (06): 150 - 153.doi: 10.13873/j.1000 - 9787 (2023) 06
- 0150 - 04.
[39] Haokun, dengchaoshuo, Zhaolu, et al. Robot path planning based on region search particle swarm
optimization [j]. Journal of electronic measurement and instrumentation, 2022, 36 (12): 126 - 135.doi:
10.13382/j. jemi B2205754.

373