Problem Research on the Optimal Way of UAV on Account of Ant

Colony Algorithm

Jichao Wang

, Meili Zhang

2,*

, Rui Tan

and Yong Xiao

Cadet Brigade 3, Dalian Naval Academy, Dalian, China

Department of Basic, Dalian Naval Academy, Dalian, China

Keywords: Sealing and Control Operations, Optimal Path, Ant Colony Algorithm.

Abstract: The sealing and control combat planning of the port is to protect our maritime transport line and lifeline. By

increasing the possibility of discovering hostile ships, it can reduce the threat of hostile forces to our

interests. It is of great significance to the security and stability of the port, and even related to the important

security of the port in the state of battle. This paper studies and discusses the function of UAV target

recognition, and proposes the optimal path problem of using the positive feedback and robustness.

1 INTRODUCTION

The identification of the merchant ships coming to

the port is an important part of the port containment

operations. At present, for the optimal path problem

of UAV (

Li Yongxia, 2016), we can use the model of

ant colony algorithm, so as to build the path

planning model built on ant colony algorithm.

Specifically, it takes advantage of the good

parallelism and simple structure. By comparing the

solution obtained by iteration, the more effective

solution is chosen. After repeated iterations, the

optimal solution can be obtained.

2 INTRODUCTION OF THE ANT

COLONY ALGORITHM

Ant colony algorithm (Deng Zijie, 2022) was founded

in 1991 by Italian scholar Dorigo etc. It is another

emerging heuristic search algorithm after neural

network, genetic algorithm and immune algorithm.

In the process of studying ant foraging, they found

that the behavior of individual ants was relatively

simple, but the ant colony as a whole could reflect

some intelligent behavior. For example, ant colonies

can care for the shortest path to the food source in

different surroundings. This is because the ants in

the colony can transmit information through some

information mechanism. After further research found

that the ants will release in the path of a can be

called "pheromone" material, ants in the colony of

"pheromone" perception, they will walk along the

"pheromone" concentration higher path, and each

passing ants will leave "pheromone", this forms a

similar positive feedback mechanism, so after a

period of time, the colony will along the shortest

path to the food source, as shown in figure 1.

Ants with shorter paths release more pheromones.

With the advancement of time, the accumulated

concentration of pheromones on shorter paths

gradually increases, and the number of ants choosing

this path is also increasing (

GUO Yanyong, 2016).

Eventually, the whole ant will focus on the best path

under the action of positive feedback, which

corresponds to the problem to be optimized.

Figure 1: Block diagram of the ant colony algorithm

program.

Wang, J., Zhang, M., Tan, R. and Xiao, Y.

Problem Research on the Optimal Way of UAV on Account of Ant Colony Algorithm.

DOI: 10.5220/0012273000003807

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 2nd International Seminar on Artiﬁcial Intelligence, Networking and Information Technology (ANIT 2023), pages 16-19

ISBN: 978-989-758-677-4

3 ESTABLISHMENT OF THE ANT

COLONY ALGORITHM

Through the ant colony algorithm, a new simulation

evolution algorithm based on the positive feedback

principle, we build the ant colony optimization

algorithm model for the optimization path problem

of UAV reconnaissance route and the target planning

problem of interception scheme (

Suwartof Basuki,

2016). As a biomimetic algorithm and universal

stochastic optimization method, it was gradually

applied to the problem of drone target interception,

after the success of the famous "Travel quotient

problem" (TSP) was achieved.

In order to solve the UAV reconnaissance route

for the classified results in the model, the shortest

path can be solved for the UAV starting from the

fixed point. The shortest flight distance of UAV is

the optimized target, and the expression is as follows

Specifically

the distance from the starting

point;

to the starting point from the starting point;

from the starting point of the first

merchant ship;

the first starting point from the

starting point;

the distance

matrix of the distance from the first merchant ship

point to the 84th merchant ship point;

the distance from the starting point from the 84th

merchant ship point.

To simulate the whole process of interception, we

need to add the following constraints

In the following, we use the classical TSP

[

JUY SUN

G Y, 2019

]

model to elaborate how to solve practical

problems based on the ant colony algorithm. For the

TSP model, in order to lose generality, let the

number of ants in the whole ant group be, the

distance between the merchant ship, and the

pheromone dimension in the connection path

between the merchant ship and the merchant ship.

At the time of beginning, the ant is

placed in diverse merchant ships, and the pheromone

concentration in the connecting way among the

merchant ships is the identical, and then the ant will

choose the route with a certain probability, or the

probability of the time ant moving from the

merchant ship to the merchant ship. Because the "ant

TSP" strategy is defined by two aspects, first the

expectation of access to an area and the

concentration of pheromone released by other ants

It is the enlightening function which

is the

expected degree of the ant moving from the

merchant ship to the merchant ship;

at the

beginning of the element.

In the process of ant traversal of various

merchant ships, similar to the actual situation, while

the ant releases the pheromones, the strength of the

pheromones in the connecting path between each

merchant ship also gradually disappears through

volatilization. To

describe this feature,

indicate the degree of pheromones. So, when all the

ants have through all the merchant ships, the

concentration of information on the connecting paths

between individual merchant ships is

Where

the pheromone concentration

increased for the release of the pheromone on the

connecting route to the

and merchant ship and the

pheromone concentration increased for the

ants.

are the general values can be calculated by the

ant week system model as

. Where the pheromone

constant indicates the total amount of pheromone

released by the ant during a cycle;

it is the

overall length of the ant's way.

For the process of solving the TSP problem

algorithm of the ant colony algorithm, the specific

meaning of each step is:

Step 1: initialization analysis of related

parameters, including the size of the colony,

pheromone factor, inspired function factor,

pheromone will send salary, pheromone in constant

maximum iteration, data into the program, the most

basic data processing, here we can generate the

distance matrix between merchant and merchant

ships.

Step 2: randomly place the drone

above the

merchant ship

, and calculate its next arrival area

for each ant, until all the ants reach all areas.

Problem Research on the Optimal Way of UAV on Account of Ant Colony Algorithm

Step 3: Calculate the path length of each ant is ,

record the optimal solution in the number of

iterations at that time, and update the pheromone

concentration on the link path of each region.

The algorithm design processes of the algorithm

are as follows:

Table 1: 84 Merchant ship coordinates for the UAV to pass

through.

Ship

serial

numbe

Initial coordinate x (in

km)

Initial coordinate y

(in km)

1 221.9130377 318.7406891

2 354.8172353 233.5129434

3 278.0278095 243.2512393

4 227.5655101 319.8244224

5 255.9664355 310.4559588

6 265.0472664 321.1379047

… … …

82 254.9713644 578.7172865

83 413.2561561 470.8937358

84 237.0798546 340.9776534

(1) Data Preparation

Clearing environment variables first is to prevent

existing variables from interfering at the same time

the merchant ship position coordinates are read from

the data file.

(2) Calculate the Merchant Ship Distance Matrix

According to the distance formula of two points in

the plane geometry, the distance between two

merchant ships is easily calculated from the

merchant coordinate matrix. However, it should be

noted that the element on the diagonal of the

calculated matrix is 0, but to ensure that the

denominator of the enlightening function is not zero,

the elements on the diagonal need to be fixed to a

sufficiently small positive number. Judging from the

order of magnitude of the data, it is enough to

correct it to the following

(3) Initialization Parameters

Before the calculation, we initialized the parameters,

and so as to speed up the execution of this program,

some processes variables involved in the program

need to be pre-allocated.

(4) Find to the Best Path Iteratively

This step is the core of the entire algorithm. Build,

first of all, according to the transfer probability of

the ant solution space, namely each ant each

merchant access, until all the merchant ship, and

then calculate the length of the ants through the path,

and update the merchant link after each iteration

pheromone update pheromone consensus path

pheromone concentration, after cycle iteration,

record the optimal path and length.

4 MODEL SOLUTION AND

ANALYSIS

To facilitate the analysis and research, on the one

hand, we show the calculation results digital or

graphically; on the other hand, we display the data

that can display the optimization process of the

program to directly present the optimization track of

the program. Through the use of Matlab's program,

the shortest path can find the shortest distance in the

process of program optimization, and the results of

its operation are as follows.

Figure 2: The best driving route of the drone.

Figure 2 shows the best route for the UAV to

experience all merchant ships. According to the

route, the UAV can take the shortest time and

experience the shortest path.

As shown in Figure 3, the UAV should start from

the 56th merchant ship position, and according to the

optimum solution of the ant colony algorithm, then

pass through 50,53... to 28, which can make the

UAV travel time and the shortest path.

ANIT 2023 - The International Seminar on Artiﬁcial Intelligence, Networking and Information Technology

Figure 3: Driving path of the drone.

As shown in Figure 4, the red line shows the

shortest distance that can be reached by the ant

colony algorithm after 200 iterations, around

3300km.

Figure 4: Ant colony algorithm results on the mean

distance and the shortest distance.

Through the above research and demonstration,

we first use the KNN algorithm to classify the three

types of merchant ships, and obtain the location of

the center point. After the classification training, its

accuracy is relatively high. Secondly, in order to

solve the UAV reconnaissance route after the

classified results in the model, we use the ant colony

algorithm to solve the best driving route from the

fixed point. From the comparison of algorithm

performance, compared with the genetic algorithm,

when the number of merchant ships is small and the

distance is close, both the ant colony algorithm and

the genetic algorithm can find the optimal solution,

and the convergence rate of the ant colony algorithm

is fast. When the quantity of merchant ships is large

and the distance is far, the ant colony algorithm can

still find the optimal solution, while the genetic

algorithm has no optimal solution, the ant colony

algorithm often takes the global optimal comparing

with other heuristic algorithms as the goal to solve

the problem, with good robustness, many

improvement directions, and is suitable for the

solution of various problems.

ACKNOWLEDGMENTS

This paper is one of the stage achievements of the

Research Fund project of Dalian Naval Academy of

the People's Liberation Army Navy (DJYKKT2023).

REFERENCES

Li Yongxia. Adaptive ant colony optimization algorithm

[D]. Chongqing University, 2016

Deng Zijie,Zeng Chuanhua, Chai Li, Shen Hangyu. Poway

optimization of rescue vehicles based on an improved

ant colony algorithm in disaster cases [D]. Xinhua

University, 2022.

GUO Yanyong,TAREK Sayed, MOHAMED H Zaki. Sa

fete evaluation of unconventional outside left- turn lane

using automated traffic conflict techniques [J].

Canadian Journal of Civil Engineering, 2016, 43 (7):

631-642.

Suwartof, Basuki K H. The application of traffic conflict

technique as a road safety evaluation method: A case

study of Hasselt intersection[J]. Applied Mechanics

and Materials, 2016, 845:394-403.

Juy, Sun G Y, Chen QH. A model combining conv-

solutional neural network and light GBM algorithm for

ultra- short-term wind power forecasting [J]. IEEE

Access, 2019 (2): 28309-28318.

Problem Research on the Optimal Way of UAV on Account of Ant Colony Algorithm