sensors

Article
Motion Control System for USV Target Point Convergence
Jian Zhou , Hui Zhang *, Kai Liu , Linhan Ma, Yanxia Yang and Zhanchao Fan

School of Information and Automation Engineering, Qilu University of Technology, Shandong Academy of
Sciences, Jinan 250300, China; 18853104864@163.com (J.Z.); liukaiyy2017@163.com (K.L.);
mlinhan906@163.com (L.M.); yangyanxia0711@163.com (Y.Y.); bhrqne5i5j@163.com (Z.F.)
* Correspondence: zhanghuifly111@126.com

Abstract: The goal of this paper is to establish a motion control system for unmanned surface vehicles
(USVs) that enables point-to-point tracking and dynamic positioning. This includes the heading
control and path following control of USVs. A hardware and software platform for USVs using
microcontrollers is designed. This paper presents the construction of a kinematics and dynamics
model for an unmanned catamaran. The motion process is divided into two segments. In the
target point tracking segment, the heading coordinate system and the ship coordinate system are
established. Based on these, a control method using differential steering to track the desired yaw
angle is designed to improve the tracking efficiency. And the accuracy of heading keeping and path
following is improved by combining the cascade PID controller. In the dynamic positioning segment,
a self-adjusting mechanism is designed, thereby enhancing the flexibility of thrust distribution and
improving the accuracy of the USV’s positioning retention in wind and wave environments. Finally,
experimental validation is carried out to verify the effectiveness of the design proposed in this paper
by issuing control commands and saving the return data through the upper computer, and then
analyzing the return data with MATLAB (R2022b, MathWorks, Natick, MA, USA).

Keywords: unmanned surface vehicles; point-to-point tracking; heading control; path following;
dynamic positioning

1. Introduction
Citation: Zhou, J.; Zhang, H.; Liu, K.; In recent years, the application of USVs in ocean space has received increasing atten-
Ma, L.; Yang, Y.; Fan, Z. Motion tion [1–3]. A USV is an intelligent, multifunctional surface platform controlled by remote
Control System for USV Target Point control or autonomous navigation [4]. It plays an important role in the fields of marine
Convergence. Sensors 2024, 24, 6589. resources exploration, maritime patrol, and meteorological observation. In order to en-
https://doi.org/10.3390/s24206589 hance the sustainable development of marine resources through the utilization of USVs,
Academic Editor: Paolo Mercorelli
research on the control of USVs is particularly important [5]. Currently, the motion research
of USVs includes heading control [6], path following control [7], and trajectory tracking
Received: 25 July 2024 control [8], among which, the correlation between heading control and path following is
Revised: 23 September 2024 more widely studied.
Accepted: 3 October 2024 The heading control of a USV is designed to rapidly change the actual yaw angle to
Published: 12 October 2024 the reference yaw angle and to minimize the amount of overshoot. In previous research,
PID controllers were first proposed and applied to real ship heading control [9]. However,
in simulations and field experiments, the USV is susceptible to model perturbations and
Copyright: © 2024 by the authors.
environmental disturbing forces, making it difficult for the PID controller to maintain
Licensee MDPI, Basel, Switzerland. a consistent control performance, and parameters need to be readjusted to stabilize the
This article is an open access article system [10,11]. Therefore, a control method with good robustness and adaptability is
distributed under the terms and needed. Wu et al. [12] added a tracking differentiator to a model-free adaptive controller
conditions of the Creative Commons and designed a controller for complex systems such as nonlinear, time-delayed, time-
Attribution (CC BY) license (https:// varying, and strongly coupled. Bu et al. [13] considered the data-driven control (DDC)
creativecommons.org/licenses/by/ problem for a class of non-affine nonlinear systems with output saturation based on model-
4.0/). free adaptive control theory. In 2018, Li et al. [14] designed a PID-MFA cascade controller

Sensors 2024, 24, 6589. https://doi.org/10.3390/s24206589 https://www.mdpi.com/journal/sensors

Sensors 2024, 24, 6589 2 of 26

based on model-free adaptive control theory to realize the heading control of USVs through
angular velocity guidance. In 2019, to address the problem that the USV heading subsystem
does not satisfy the quasi-linear assumption, Liao et al. [15] redefined the output of the USV
as a linear sum of heading and angular velocity, and based on this redefinition proposed a
free adaptive control method for the compact format model. However, the above methods
are more dependent on the accuracy of the motion model, complicated in design, have
high environmental uncertainty, and have not been applied in real environments. Based on
the heading coordinate system and ship coordinate system, this paper designs a method of
tracking and maintaining the desired heading yaw using differential speed control, and
establishes a cascade PID controller to optimize the control accuracy. Finally, it is practically
applied to USVs and achieves a good heading control effect.
Path following is one of the fundamental motion control problems in the autonomous
control of USVs. Path following is defined as a USV following a time-independent path at
a certain speed [16]. Currently, the well-known bootstrap method is LOS (Line of Sight)
bootstrap [17]. To address external marine environmental disturbances such as constant or
time-varying winds, waves, and currents, Fossen et al. proposed ILOS (Integral LOS) and
ALOS (Adaptive LOS) based on proportional LOS guidance methods [18]. As the core of
the path-following control system, the control algorithm is the key to determining whether
the USV can accomplish the intended task.
A system that maintains a surface vehicle in a predetermined position and heading
by means of an appropriate propeller and propeller action is known as a dynamic po-
sitioning (DP) system. The technology is applicable to ships, which can use their own
propulsion instead of traditional mooring systems to maintain the desired position and
heading in the event of disturbances on the water [19], and the problem of controlling the
dynamic positioning system of ships has been a hot topic of research and discussion among
international experts and scholars [20]. Currently, DP has developed into an important
technology for deep-sea resource exploration and is widely used on drill ships, cable-laying
vessels, supply vessels, survey vessels, and offshore platforms [21]. This emphasizes the
considerable theoretical importance and practical utility of DP [22]. However, the dynamic
positioning system is usually applied to large ships, while the application in small USVs is
less common. One of the research objectives of this paper is to design a simple and efficient
dynamic positioning device that can be practically applied to small USVs, which can help
small USVs keep in the predetermined position for operation and reduce the interference
of wind and waves on the vehicles.
The rest of the paper is as follows. Section 2 designs the general structure of the USV.
Section 3 models the dual-propeller differential speed USV to provide theoretical support
for the later scheme. Section 4 proposes a solution for heading tracking and keeping, and
Section 5 describes the working principle of the adaptive adjustment mechanism and carries
out real ship experiments to verify the above solution. Finally, the paper is summarized in
Section 6.

2. Overall Design of the USV

2.1. Hull Structure
The overall design of the USV is the prerequisite and foundation for researching the
design of autonomous navigation control for USVs. The hull structure of the USV employed
in this paper is a dual-propeller catamaran structure. The control system of the USV is
mainly comprised of two parts: the ship-based system and the shore-based system.
As shown in Figure 1, this paper enhances the traditional catamaran structure. In
the traditional catamaran, the propeller is typically fixed at the rear of the hull. The
disadvantage of this structure is that it can only exert longitudinal force on the hull. To
overcome this problem, this paper employs fixed brackets to connect the propeller to
waterproof steering gears and positions them on both sides of the USV. The steering gear
rotates to drive the propellers to rotate, enabling the thrust generated by the propellers to
be distributed in all directions. The ship-based system is located in the control box.
Sensors 2024, 24, 6589 3 of 26

Figure 1. Physical and model drawing of hull structure.

2.2. Overall Design of Ship-Based System

As shown in Figure 2, the ship-based control system is composed of four main compo-
nents, namely ship-based main controller, positioning system, remote control system, and
propulsion system.

Figure 2. Design of ship-based control system.

The design idea of the positioning system is to utilize the Global Positioning Sys-
tem (GPS) mobile station to acquire the latitude, longitude, real-time speed, and other
information of the USV. Subsequently, through the Inertial Measurement Unit (IMU), the
yaw angle and steering angular velocity of the USV’s bow, among other information, are
obtained. The aforementioned information is transmitted to the ship-based main controller
via 232 communication and 485 communication, respectively. After analysis by the main
controller, the position and attitude information of the USV can be obtained.
The design of the remote control system is to use the transmitter to output control
instructions. According to the control instructions, the transmitter sends data to the
receiver. Then, through the SBUS to serial module, the data are sent to the ship-based
main controller. The main controller parses the data and finally obtains the control code to
control the propellers.
Sensors 2024, 24, 6589 4 of 26

The design idea of the propulsion system is to utilize electronic speed controls to
control the brushless motors to drive the movement of the USV. In the target point tracking
segment, the dual-propellers USV employs the rotational speed difference to turn, which
significantly reduces the turning radius and enhances flexibility. Due to the real-time
acquisition and processing of information, when the desired yaw angle information is
received, the main controller outputs PWM waves to control the propellers in real time.
Moreover, the dual-propellers USV saves the time by steering through non-essential paths
and is capable of adjusting to the desired yaw angle more quickly, thus shortening the
response time. In the dynamic positioning segment, the position of the USV is corrected
through the cooperation of the steering gears and propellers. The system is powered by
lithium battery packs.

2.3. Overall Design of Shore-Based System

As shown in Figure 3, the design idea of the shore-based system is that the main con-
troller of the shore-based system receives latitude, longitude, and target point parameters
information set by the upper computer through serial communication. Then, it sends this
information to the ship-based main controller via wireless communication. The ship-based
control system receives the information and controls the USV to navigate autonomously.
Simultaneously, real-time latitude, real-time longitude, real-time yaw angle, and other
information are also transmitted back to the main controller of the shore base via wireless
communication, Subsequently, the information is visualized and stored through the upper
computer. The GPS mobile station utilizes the differential data sent by the GPS base station
to perform RTK accurate positioning.

Figure 3. Design of shore-based control system.

3. Mathematical Modeling of USV

3.1. Establishment of Coordinate Systems
The motion of the USV is classified as a six-degree-of-freedom state of motion, which
is generally described based on two coordinate systems: a North-East-Down coordinate
system, with the center of the Earth as the origin of the coordinates, and a body-fixed
coordinate system, with the USV itself as the origin of the coordinates.
As shown in Figure 4, in the North-East-Down coordinate system OXYZ, the X-axis
points north, the Y-axis points due east, and the Z-axis points to the center of the earth. The
coordinate system O1 X1 Y1 Z1 refers to the body-fixed coordinate system, with the X1 -axis
pointing in the direction of the ship’s bow, the Y1 -axis perpendicular to the X1 -axis pointing
to the right side of the ship, and the Z1 -axis perpendicular to the X1 O1 Y1 plane pointing
to the center of the earth. The physical quantities of the body-fixed coordinate system are
shown in Table 1.
Sensors 2024, 24, 6589 5 of 26

Figure 4. Schematic diagram of the coordinate system.

Table 1. Representation of physical quantities in the coordinate system.

Serial Force (N)/ Linear Velocity (m/s)/ Position/

Description
Number Moment (N·m) Angular Velocity (rad/s) Attitude Angle (°)
1 surge X u x
2 sway Y v y
3 heave Z w z
4 roll K p ϕ
5 pitch M q θ
6 yaw N r ψ

3.2. Kinematic Model

We generally ignore the effects of heave, roll, and pitch motions. The six-degree-of-
freedom motion is simplified to a three-degree-of-freedom motion of surge, sway, and
yaw. From this, combining Equations (1) and (2) can be simplified to obtain a simplified
mathematical model:
    
ẋ cosθcosψ sinϕsinθcosψ − cosϕsinψ cosϕsinθcosψ + sinϕsinψ u
 ẏ  =  cosθsinψ sinϕsinθsinψ + cosϕcosψ cosϕsinθsinψ − sinϕsinψ  ν  (1)
ż − sin θ sinϕcosθ cosϕcosθ w
    
ϕ 1 sinϕtanθ cosϕtanθ p
 θ̇  = 0 cosϕ −sinϕ  q  (2)
ψ̇ 0 sinϕsecθ cosϕsecθ r

 ẋ = u cos ψ − ν sin ψ

ẏ = u sin ψ + ν cos ψ (3)

ψ̇ = r


3.3. Dynamics Model

The position and attitude angle changes of a USV under the influence of external
forces and moments are illustrated by means of a dynamics model. To facilitate our analysis
of the model, the following assumptions are made:
(1) Since the USV is moving in the horizontal plane, ignoring the effects of surge, sway,
and yaw motions, z = 0, w = 0, ϕ = 0, p = 0, θ = 0, q = 0.
(2) The mass of the hull is evenly distributed and symmetrical, so the values of Ixy and
Iyz are both 0.
(3) The origin of the body-fixed coordinate system is set at the center of gravity of the
USV so that the center of gravity coordinates x1g , y1g , and z1g are all 0.
Sensors 2024, 24, 6589 6 of 26

Based on the above assumptions, a mathematical model of the horizontal motion of

the USV can be constructed.

M ν̇ + C (ν)ν + D (ν)ν = τ + τE (4)

In Equation (4), v = [u, v, r ] T , τ = [τu , τv , τr ] T , and τu , τv , and τr are the ship’s

longitudinal thrust, transverse thrust and turning moment, respectively. In particular, it is
noted that in the target point tracking segment τv = 0 at this time, since the thrust only acts
in the longitudinal direction of the USV. In the dynamic positioning segment, the USV is no
longer steered by differential speed, so at this point τr = 0, τE = ⌈τE,u , τE,v , τE,r ⌉ T denotes
the external environmental perturbation. The specific expressions of each matrix and the
meanings they represent are explained below:
(1) Inertia matrix
 
m11 0 0
M= 0 m22 m23  (5)
0 m23 m33

M represents the inertia matrix of the system, which is a positive definite matrix,
m11 = m − Xu̇ , m22 = m − Yv̇ , m33 = Iz − Nṙ . m denotes mass, Xu̇ , Yv̇ , and Nṙ denote
longitudinal, transverse, and navigational additional mass and moment coefficients of
inertia, and Iz denotes moment of inertia.
(2) Coriolis matrix
 
0 0 −m22 ν − m23 r
C (ν) =  0 0 m11 u  (6)
m22 ν + m23 r −m11 u 0

C (v) reflects the effects that a ship undergoes during a particular rotational motion at
sea, taking into account the role of additional mass. Although this matrix does not directly
represent the actual forces and moments exerted on the ship, its influence is critical in the
study of ship motion in a non-inertial reference system.
(3) Hydrodynamic damping coefficient matrix
 
d11 0 0
D (ν) =  0 d22 d23  (7)
0 d32 d33

Under ideal fluid conditions, D(v) exhibits asymmetric and strictly positive definite
properties. d11 = −Xu , d22 = −Yv , d33 = − Nr . Xu , Yv , and Nr are linear damping coefficients.
Most ships have left–right symmetry and move in a two-dimensional plane. In the
ideal situation, the USV is assumed to be a rigid body so that the interaction between
each mass element is neglected. To simplify the model and consider the properties of
the inertia matrix, M, and the damping matrix, D, at low speeds, Fossen [23] noted that
the non-diagonal elements were insignificant compared with the diagonal elements, and
therefore ignored the influence of these non-diagonal elements.
 
m11 0 0
M=  0 m22 0  (8)
0 0 m33
 
0 0 −m22 ν
C (ν) =  0 0 m11 u  (9)
m22 ν −m11 u 0
Sensors 2024, 24, 6589 7 of 26

 
d11 0 0
D (ν) = 0
 d22 0  (10)
0 0 d33
The simplified equations for the kinematics and dynamics of the USV can be expressed as:


 ẋ = u cos ψ − ν sin ψ

ẏ = u sin ψ + ν cos ψ





ψ̇ = r




m22 d τu + τE,u

vr − 11 u +

u̇ =

m 11 m 11 m11 (11)

m d22 τv + τE,v


v̇ = − 11 ur −


 v+
m m m11

22 22




ṙ = m11 − m22 uν − d33 r + τr + τE,r



m33 m33 m33

Considering that the catamaran, which has a uniform mass distribution, is chosen as
the main body of the USV in this paper, we simplify the hull as a mass point in performing
the longitudinal force analysis. In this case, the combined force of the propulsive forces in
the longitudinal and transverse directions are expressed as:

τu = ( Fl + Fr ) · sin φ (12)

τv = ( Fl + Fr ) · cos φ (13)
Fl represents the effective thrust of the left propeller and Fr represents the effective
thrust of the right propeller. φ is the angle between the line where the combined force of
the propellers is located and the line where O1 Y1 is located in the body-fixed coordinate
system. The value of φ is in the range [0, 90].
In the target point tracking segment, the propellers’ thrust always acts on the longitu-
dinal direction of the catamaran, so φ = 90, τu = Fl + Fr , τv = 0. In this case, the expression
for the moment produced by the propellers is:

τr = ( Fl − Fr ) · Bd (14)

Bd represents the distance of the two propellers relative to the center axis of the hull.
In the target point tracking segment, we can further reformulate the dynamics equation
of the USV as:


 ẋ = u cos ψ − v sin ψ

ẏ = u sin ψ + v cos ψ





ψ̇ = r




m22 d F + Fr + τE,u

vr − 11 u + l

u̇ =

m11 m11 m11 (15)

m11 d22

 τE,v
v̇ = − m22 ur − m22 v + m22







ṙ = m11 − m22 uν − d33 r + ( Fl − Fr ) · Bd + τE,r



m33 m33 m33

In the dynamic positioning segment, since the direction and magnitude of thrust
output from both sides of the propellers remain the same, Fl = Fr , and according to
Sensors 2024, 24, 6589 8 of 26

Equation (14), we can obtain τr = 0. We can reformulate the dynamic equation of the USV
as follows: 


 ẋ = u cos ψ − v sin ψ

ẏ = u sin ψ + v cos ψ





ψ̇ = r




m22 d ( F + Fr ) · sin φ + τE,u

vr − 11 u + l

u̇ =

m11 m11 m11 (16)

( F + Fr ) · cos φ + τE,v

 m d22
v̇ = − 11 ur − v+ l



m m m22

22 22




ṙ = m11 − m22 uν − d33 r + τE,r




m33 m33 m33

3.4. Parameter Identification of Model

The parameter identification of the model of the target point tracking segment is
carried out using the uniform linear motion experiment and the slewing motion experiment.
Table 2 shows the basic parameters of the USV.

Table 2. The basic parameters of the USV.

Parameter Value
Quantity m/kg 32.5
Length L/m 1.05
Height B/m 0.8
Draught depth at full load D/m 0.17
Distance from propeller to centerline Bd /m 0.42

Based on the above parameters, it can be calculated that m11 = 34.64, m22 = 60.02,
m33 = 10.24. Next, the parameter identification of the coefficient matrix, D, is carried out.
d11 = ∂τ u ∂τv ∂τr
∂u , d22 = ∂v , d33 = ∂r .
In the control system designed in this paper, the control code n (0–700),which corre-
sponds to the input control voltage of the propeller (0–4 V), is set so as to control the output
power of the propeller, and the linear relationship between nl and the voltage, Vl , for a
single propeller is:
Vl = 0.0057nl (17)
According to the propeller manual, the relationship between individual propeller
thrust and controlled voltage is as follows:

Fl = 9.5Vl (18)

So the combined force of the dual propellers in the forward direction of the USV is
F = τu = 9.5(Vl + Vr ), and the moment τr = 9.5Bd (Vl − Vr ). So, given a control code, the
thrust of the propeller can be obtained, and then the longitudinal or angular velocities
produced by different thrusts can be measured. It is then possible to fit a relationship
between thrust and velocity or angular velocity.
Substituting Equation (17) into Equation (18), the relationship between the combined
force and the control code when advancing is obtained as follows:

F = 0.1083nl (19)

The experiment of uniform linear motion with different speeds was designed, and
the real-time data were collected using the host computer to obtain the correspondence
between the combined force and the current speed, and the experimental data are shown
in Table 3. The fitting of the combined force and velocity is shown in Figure 5.
Sensors 2024, 24, 6589 9 of 26

Table 3. Velocity values under different uniform motions.

Control Code nl Control Code nr Combined Force F/N Current Velocity/(m/s)

100 100 10.83 0.044
150 150 16.245 0.091
200 200 21.66 0.267
250 250 27.075 0.344
300 300 32.49 0.492
350 350 37.905 0.562
400 400 43.32 0.623
450 450 48.735 0.742

Figure 5. Linear relationship between velocity and combined force.

Linear relationship after fitting: F = 52.37u + 9.1153. Therefore, d11 = ∂τu ∂F

∂u = ∂u = 52.37.
Due to the experiment in a still-water lake, it is assumed that the lateral thrust of the
USV in the tracking segment at the target point is negligible and the lateral velocity is 0.
Therefore d22 = 0, v = 0.
Combining Equations (17) and (18), the relationship between control code and moment
can be shown as follows:
Fr = 0.0227(nl − nr ) (20)
Slewing experiments with different speeds were designed to obtain the relationship
between moment and angular velocity, as shown in Table 4. The moments and angular
velocities were fitted as shown in Figure 6. Therefore, d33 = ∂τr ∂Fr
∂r = ∂r = 0.1875.

Figure 6. Linear relationship between angular velocity and moment.

Sensors 2024, 24, 6589 10 of 26

Table 4. Moment values and angular velocity at different speeds.

Control Code nl Control Code nr Moment Fr /N·m Angular Velocity/(°/s)

150 100 1.135 2.085
170 100 1.589 3.791
190 100 2.043 5.494
210 100 2.497 7.558
230 100 2.951 10.748
250 100 3.405 13.429
270 100 3.859 15.623
300 100 4.54 19.742

In summary, by substituting the values of d11 , d22 , d33 , m11 , m22 , m33 into Equation (15)
and ignoring the external perturbation, τE , the mathematical model of USV motion can be
obtained as follows: 

 ẋ = u cos ψ − v sin ψ

ẏ = u sin ψ + v cos ψ



ψ̇ = r (21)

u̇ = −1.51u + +0.029( Fl + Fr )





ṙ = −0.018r + 0.041( Fl − Fr )


4. Target Point Tracking

During target point tracking, the bow angle of the USV initially needs to track the
desired yaw angle and maintain it. In this segment, the output thrust of the propellers is
always in the longitudinal direction of the hull, thereby satisfying the catamaran motion
model described above. In this paper, a heading coordinate system is designed to calculate
the desired yaw angle in real time by utilizing the position changes of the USV to always
track the direction of the target point. The actual yaw angle is compared with the desired
yaw angle to determine the working conditions of the propellers on both sides. Then, the
angular cascade PID controller is utilized to adjust the rotational speed of the propellers on
both sides to achieve the purpose of fast and accurate tracking of the desired yaw angle.
Finally, a positional cascade PID controller is utilized to enable the USV to rapidly converge
on the target position.

4.1. Establishment of the Heading Coordinate System

As shown in Figure 7, a two-dimensional heading coordinate system, OXY, is estab-
lished. In this coordinate system, the direction of due north is taken as the positive direction
of the X-axis and the direction of due east is taken as the positive direction of the Y-axis.
The target point is set as the origin, O.
The latitude and longitude are transformed to planar coordinates through Mercator
transformation. The coordinates of the target point are set as ( xd , yd ) , The coordinates of
the current position of the USV are set to ( x, y). The longitudinal error, xe , and lateral error,
ye , in the heading coordinate system are shown below:
(
xe = x − xd
(22)
ye = y − yd

Based on the positivity and negativity of xe and ye , we specify in which quadrant the
USV is located at the target point. Quadrant ⃝
1⃝2⃝ 3⃝4 is as follows:
( ( ( (
xe > 0 xe > 0 xe < 0 xe < 0
⃝1 ⃝
2 ⃝
3 ⃝4 (23)
ye > 0 ye < 0 ye < 0 ye > 0
Sensors 2024, 24, 6589 11 of 26

Figure 7. Schematic diagram of the heading coordinate system.

As shown in Figure 7, ψc is the azimuthal yaw angle of the USV. After judging the
quadrant where the USV is located, the angle values of ψc in different quadrants are
calculated, respectively, and the desired yaw angle, ψd , is finally obtained. The formula is
as follows:
|x |

 arctan( |ye | ) × 180
e
ψc = −90 +



 ⃝
1
 π
|x |

arctan( |ye | ) × 180



e

ψc = 90 −

⃝
2
π (24)
|x |

 arctan( |ye | ) × 180
e
ψc = 90 − ⃝
3



 π
 |x |
arctan( |ye | ) × 180


 e
ψ = 90 + ⃝
4

c
π
(
ψd = ψc − 180 ψc > 0
(25)
ψd = ψc + 180 ψc ≤ 0
A two-dimensional ship coordinate system, O2 X2 Y2 , is established, taking due north
as the X2 -axis positive direction, due east as the Y2 -axis positive direction, and the center of
the USV as the origin, O2 . In Figure 7, ψi represents the actual yaw angle of the USV, which
is the angle in the direction of the bow, and ψi is obtained by IMU module after processing.
By combined Figure 7, and Equations (24) and (25) to analyze, the value ranges of the
azimuthal yaw angle, ψc , desired yaw angle, ψd , and actual yaw angle, ψi , are the same.
All of them have a due north direction of 0, a clockwise rotation angle range of (0, −180),
and a counterclockwise rotation angle range of [0, 180]. Therefore, as long as the values of
ψi and ψd are equal, this indicates that the bow of the USV is pointing at the target point at
this moment. In Figure 7, d is the distance between the USV and the target point, which is
calculated as follows: q
d = xe2 + y2e (26)

4.2. Design of Heading Tracking Controller

4.2.1. Principles of Automatic Heading Control
The design idea of the heading tracking controller is to steer the USV by controlling
the output moment of the propellers. By comparing the actual yaw angle, ψi , the desired
yaw angle, ψd , and the azimuthal yaw angle, ψc , the operating status of the left propeller,
ML , and the right propeller, MR , are determined. The operating state of the propeller
Sensors 2024, 24, 6589 12 of 26

determines whether the USV turns clockwise or counterclockwise, ultimately achieving

the goal of turning the smallest possible angle to be able to track the desired heading of
the USV.
As shown in Figure 8, it is assumed that the USV is in the first quadrant, and its actual
position remains unchanged during the steering process. Additionally, the position of
the target point is fixed. According to Equations (24) and (25), it can be analyzed that the
values of ψc and ψd remain unchanged. The USV regulates the actual yaw angle through
differential steering of the propellers, so only the value of ψi changes. Since the value range
of the angles in the heading coordinate system and the ship coordinate system are the
same, the angular representation in the heading coordinate system can be discussed in the
ship coordinate system. The two cases in Figure 8a are used as examples to analyze the
operating state of the propellers when tracking the desired yaw angle.

(a) Analysis of the situation when ψi >0 in quadrant ⃝

1

(b) Analysis of the situation when ψi <0 in quadrant ⃝

1

Figure 8. The principle of heading tracking.

As shown in Figure 8a, when the value of the actual yaw angle, ψi , is greater than 0.
If ψi is less than ψd , the right propeller, MR , should rotate forward and the left propeller,
ML , should rotate backward. This is to push the USV to turn counterclockwise so that ψi
converges to ψd . If ψi is greater than ψd , then the left propeller, ML , rotates forward and
the right propeller, MR , rotates backward, pushing the USV clockwise so that ψi converges
to ψd .
As shown in Figure 8b, when the value of the actual yaw angle ψi is less than 0. If
ψi is greater than ψc , the right propeller, MR , should rotate forward and the left propeller,
ML , should rotate backward. This is to push the USV to turn counterclockwise so that ψi
converges to ψd . If ψi is less than ψc , the left propeller, ML , rotates forward and the right
Sensors 2024, 24, 6589 13 of 26

propeller, MR , rotates backward, pushing the USV to turn clockwise so that ψi converges
to ψd .
The control shown in Figure 9 only determines whether the propellers are operating
in forward or reverse rotation and does not precisely control the rotational speed of the
propellers. Therefore, a faster response bang-bang control is employed. In Figure 9, FWD
indicates forward rotation and REV indicates reverse rotation. Regardless of the direction
in which the USV is located with respect to the target point, the judgment principle shown
in Figure 9 is used to quickly determine how the left propeller and the right propeller
should work so that the USV can turn at the smallest possible angle to be able to track the
desired yaw angle, thereby improving the response speed.

Figure 9. Propeller operations on both sides when tracking the desired yaw angle.

4.2.2. Design of Heading Cascade PID Controllers

The USV not only needs to rapidly converge to the desired heading but also needs
to accurately track the desired heading. When navigating in real water environments, the
USV is bound to be affected by wind and waves. Thus, the USV must be able to complete
target point tracking by adjusting its heading in real time when it is disturbed. In this paper,
cascade PID controllers are employed to control the yaw angle and target point distance.
Figure 10 shows the process diagram of the angular cascade PID controller. In the
process of tracking the target point, the USV first tracks the desired heading through
steering. During the steering process, the USV measures the current actual yaw angle, ψi ,
in real time via the IMU module. At the same time, it calculates the desired yaw angle, ψd ,
based on the relative position of the USV and the target point. Finally, it obtains the yaw
angle deviation, ∆ψ, which is calculated by the following formula:

∆ = |ψd − ψi | (27)
(
∆ψ = ∆ ∆ ≤ 180
(28)
∆ψ = 360 − ∆ ∆ > 180
This deviation value is fed into the PID controller of the outer loop, and the output
equation of the outer loop PID controller is:

d∆ψ
Z
rd = K po1 ∆ψ + Kio1 ∆ψdt + Kdo1 (29)
dt
Sensors 2024, 24, 6589 14 of 26

Figure 10. Angular cascade PID controller.

The outer loop controller is calculated to output a desired angular velocity value, rd .
K po1 , Kio1 , and Kdo1 are the proportional, integral, and differential coefficients, respectively,
of the outer loop of the angular cascade PID controller.
The IMU module of the USV also measures the current actual angular velocity, ri , and
compares it with the output of the outer loop to obtain the yaw angular velocity deviation
∆r = rd − ri . This deviation value is fed into the PID controller of the inner loop, which is
given by the output equation of the inner loop PID controller:

d∆r
Z
Ca = K pi1 ∆r + Kii1 ∆rdt + Kdi1 (30)
dt
After calculation, the inner loop controller outputs the control signal, Ca . This control
signal then generates the PWM to control the rotational speed of the propeller during
steering. K pi1 , Kii1 , and Kdi1 are, respectively, the proportionality coefficients, integra-
tion coefficients, and differential coefficients of the inner loop of the angular cascade
PID controller.
Figure 11 shows the process diagram of the positional cascade PID controller. Once
the tracking of the yaw angle is achieved, the propellers on both sides maintain the same
rotational speed to push the USV in the direction of the desired yaw angle and gradually
converge to the target point. While moving forward, the USV measures the current actual
position ( x, y) in real time via the GPS module. The desired position ( xd , yd ) is set by the
upper computer and sent to the ship-based controller through wireless communication.
The actual distance between the USV and the target point can be obtained according to
Equation (26). Let ed = d, and input ed into the outer loop PID controller with the formula:

ded
Z
ud = K po2 ed + Kio2 ed dt + Kdo2 (31)
dt
The outer loop is calculated to output the desired speed, ud . K po2 , Kio2 , and Kdo2 are
the proportional, integral, and differential coefficients, respectively, of the outer loop of the
positional cascade PID controller.

Figure 11. Positional cascade PID controller.

The GPS module simultaneously measures the actual velocity, ui , and compares it
with the output of the outer loop to obtain the velocity deviation ∆u = ud − ui . The velocity
deviation, ∆u, is input to the inner-loop PID controller, which is given by the formula:

d∆u
Z
Cd = K pi2 ∆u + Kii2 ∆udt + Kdi2 (32)
dt
Sensors 2024, 24, 6589 15 of 26

The inner loop controller, after calculation, outputs the control signal Cd , which in
turn yields the PWM to control the speed of the propeller as it advances. K pi2 , Kii2 , and Kdi2
are the proportional, integral, and differential coefficients of the inner loop of the positional
cascade PID controller, respectively.

4.2.3. Stability Analysis

Table 5 gives the parameter values of the cascade PID controllers and the values are
obtained through actual experiments.

Table 5. Parameter information of the controller.

Parameter Name Value Parameter Name Value

K po1 2.5 K po2 23
Kio1 0.03 Kio2 0.11
Kdo1 1.1 Kdo2 8.5
K pi1 1.5 K pi2 7.5
Kii1 0.01 Kii2 0.13
Kdi1 0 Kdi2 0

Combining Equations (18) and (21) yields the following equation:

(
u̇ = −1.51u + 0.2755V1
(33)
ṙ = −0.018r + 0.3895V2

where V1 = Vl + Vr , V2 = Vl − Vr , Vl , Vr represent the voltage of the left and right pro-

pellers, respectively. The transfer functions G1 (S) and G2 (S) are obtained according to
Equation (33) as follows, respectively:

0.2755

 G1 (S) =


s + 1.51
(34)
 0.3895
 G2 (S) =

s + 0.018

The transfer function, G2 (S), is used as the controlled object of the angular cascade
PID controller, so the open-loop transfer function of the angular cascade PID controller can
be obtained as:

0.642175s3 + 1.900775s2 + 0.027265s + 0.00011685

Ga (S) = Ga1 (S) Ga2 (S) G2 (S) = (35)
s2 (s + 0.018)

Ga1 (S), Ga2 (S) are the transfer functions of the angular loop cascade PID controller
and angular velocity loop cascade PID controller, respectively. The poles are found to be 0,
0, −0.018 according to the above equation. So the system is stable.
The transfer function, G1 (S), is used as the controlled object of the positional cascade
PID controller to obtain the open-loop transfer function of the system as:

17.561875s3 + 47.801375s2 + 1.050625s + 0.00391865

G p (S) = G p1 (S) G p2 (S) G1 (S) = (36)
s2 (s + 1.51)

G p1 , G p2 are the transfer functions of the position loop cascade PID controller and
velocity loop cascade PID controller, respectively. According to the above equation, the
poles are obtained as 0, 0, −1.51, so the system is stable.

5. Dynamic Positioning and Ship Experiment

The presence of various disturbing factors in the water environment will lead to the
transverse drifting movement of the hull. The USV is differentiated from large vessels in
Sensors 2024, 24, 6589 16 of 26

that it does not have direct propellers in the transverse direction of the hull, and the actuator
is prone to fall into thrust limitation. To address these issues, an adaptive adjustment
mechanism is designed in this paper. This enables the propellers on both sides of the hull to
no longer be limited to applying longitudinal thrust but to be able to flexibly distribute the
thrust in all directions to compensate for the hull displacement caused by wind and wave
currents. Finally, the whole process of the USV tracking the target point and remaining at
the target point is verified through real ship experiments, proving the validity of the design.

5.1. Adaptive Adjustment Mechanism

5.1.1. Hardware Architecture
As shown in Figure 12, the hardware component of the adaptive adjustment mecha-
nism mainly comprises steering gears, propellers, connections, and fixed supports. The
rotation of the steering gear drives the rotation of the propeller, thereby adjusting the direc-
tion of thrust and enabling the USV to output thrust in any direction. The aforementioned
program provides hardware support to overcome the transverse drift motion.

Figure 12. Hardware components of the adaptive adjustment mechanism.

5.1.2. Principle of Adaptive Adjustment

As shown in Figure 13, ψel is the rudder angle of the left steering gear of the USV, and
ψer is the rudder angle of the right steering gear, both of which are oriented with the ship’s
bow facing 0°. The angle ranges are shown in Figure 13. The blue arrows indicate the
direction of thrust generated by the forward rotation of the propellers when the steering
gears are rotated at different angles. The green arrows indicate the direction of thrust when
the propellers are reversed. Although the steering gears rotate only 180°, the propellers can
generate thrust in the opposite direction, allowing for a 360° thrust distribution. Assuming
that the yellow arrow in the figure indicates the disturbance caused by wind and wave
currents, in order to overcome the effects caused by this disturbance, the angle of the
steering gears on both sides and the forward and reverse rotations of the propellers are
controlled so that the thrust output of the USV is counteracted with the disturbance, which
is represented as blue and green arrows with a yellow outer frame in Figure 13.
As shown in Figure 14, the rudder angles are analyzed using the heading coordinate
system and the ship coordinate system in the event of transverse drift motion. If the USV
is outside the circle, it is regarded as the target point tracking segment. At this time, the
rudder angles on both sides are maintained at 180°. Regardless of the forward or reverse
rotation of the propeller, the output thrust is only in the longitudinal direction. When the
USV enters the circle, it is considered as a dynamic positioning segment. At this time, the
steering gear starts to work to compensate for the transverse drift motion by changing the
thrust direction.The following is an analysis of how the rudder angles are determined and
how the propellers operate to generate thrust to compensate for the interference.
Sensors 2024, 24, 6589 17 of 26

Figure 13. Schematic diagram of adaptive adjustment principle.

(a) Analyze rudder angles and propeller operations at ψd >0

(b) Analyze rudder angles and propeller operations at ψd <0

Figure 14. The angle of steering gears and the operation of propellers.
Sensors 2024, 24, 6589 18 of 26

As shown in Figure 14a, the position of the USV is shifted within the circle due to
interference. The desired yaw angle ψd > 0, i.e., the USV is in quadrant ⃝ 1 ⃝,
4 at which
point the relationship between ψi and the values of ψd and ψc is compared. If the value
of ψi is (ψc , ψd ), the left rudder angle, ψel , needs to be adjusted to ∆ψ, the left propeller,
ML , reversed, the right rudder angle, ψer , adjusted to 180 − ∆ψ, and the right propeller,
MR , forwarded, in order to generate thrust toward the target point. If the value of ψi
is [ψd , 180] ∪ (−180, ψc ], the left rudder angle, ψel , needs to be adjusted to 180 − ∆ψ, the
left propeller, ML , forwarded, the right rudder angle, ψer , adjusted to ∆ψ, and the right
propeller, MR , reversed, in order to generate a thrust that tends to the target point.
Rudder angles ψel and ψer on both sides satisfy equation:

ψel + ψer = 180 (37)

As shown in Figure 14, the values of the rudder angles are related to the value of
deviation angle ∆ψ between the actual yaw angle and the desired yaw angle. The angle
of the steering gear on the side where the propeller is reversed is equal to the value of
deviation angle ∆ψ, and the angle of the steering gear on the other side is 180 − ∆ψ. Based
on this logic, it is also possible to determine the rudder angles and propeller operations
when the desired yaw angle ψd < 0 in Figure 14b.
Figure 15 shows overall process for determining the operation of propellers and the
angle of steering gears in different situations of the dynamic positioning segment. The
deviation, ∆ψ, between the desired and actual yaw angle is changed in real time. The left
rudder angle, ψel , and the right rudder angle, ψer , change in real time with the yaw angle
deviation, ∆ψ, always outputting a thrust converging to the target point and constantly
correcting the position of the USV. The above control is real time in nature. Accordingly,
the position offset of the USV due to environmental perturbations can be minimized, and
the accuracy of the positioning hold is improved.

(a) Flowchart for determining the rudder angles and the operation of the propellers

(b) Adaptive regulation process

Figure 15. Overall design of the adaptive regulation process.

Sensors 2024, 24, 6589 19 of 26

5.2. Experiment
In order to verify the application of the methodology used in this paper in real
navigation control, navigation experiments were carried out on a lake within the Garden
Expo Park in Changqing District, Jinan City. The data obtained from the USV were
transmitted to the upper computer through the wireless module to display and create
an Excel table to store the data. Figures 16 and 17 show the experimental real picture and
the schematic diagram of the upper computer.

Figure 16. Schematic diagram of the experimental environment of the USV.

Figure 17. Upper computer information display.

Sensors 2024, 24, 6589 20 of 26

After processing the redundant and anomalous data in the Excel sheet, the data in the
sheet are plotted and analyzed by MATLAB.
In addition to analyzing the navigational data acquired by the USV as described above,
it is also important to analyze the disturbances. Since it is difficult to observe the wind
and waves directly, we use numerical simulation to estimate and measure the disturbance
information. Since the experimental environment is a still-water lake, we focus on the
effects of wind and current on the navigation experiments to give the basic model for
calculating the wind interference force. The force and moment of the steady wind with
direction αw and velocity Uw on the USV are expressed as:

Xw = − 21 ρ a A T u2rw + v2rw Cwx (αw )






Yw = 12 ρ a A L u2rw + v2rw Cwy (αw )


(38)


 N = 1 ρ A L u2 + v2 C ( α )


w 2 a L rw rw wn w

where urw and vrw are the speed of the ship after being affected by the wind, ρ a is the air
density, A T and A L are the orthographic projection and measured projection areas above
the USV’s waterline, respectively, and Cwx , Cwy , and Cwn are the wind coefficients in the X
and Y directions and the wind moment coefficient around the Z-axis, respectively.
Assuming that the direction and velocity of the uniform current are αwa and Uwa , the
relative speed of the USV can be obtained as:
(
urwa = u + Uwa cos(αwa − ψ)
(39)
vrwa = v + Uwa sin(αwa − ψ)

Table 6 gives the meteorological information and the basic parameters of the USV on
the day of the experiment:

Table 6. Basic parameter information.

Date of experiment 20 May 2024 Length of USV (m) 1.05

The direction of the wind southwestern Horizontal projected area (m2 ) 0.15
The speed of the wind (Kt) 7.5 Vertical projected area (m2 ) 0.072

5.2.1. Data Analysis of Target Point Tracking Segment

As shown in Figure 18, the latitude and longitude are converted to planar coordinates
for analysis after Mercator transformation. The horizontal coordinates are the transformed
coordinates of the longitude information, and the vertical coordinates signify the trans-
formed coordinates of the latitude information. The positive direction of the Y-axis is
northward. In this paper, curved path following is not explored. Hence, the starting
point and the goal point are connected by a straight line, which serves as the desired path.
Initially, the USV tracks the desired yaw angle by differential steering, resulting in the USV
moving irregularly around the starting point. After the steering is completed, the USV
follows the desired path. However, there is uncertainty regarding the cause of the USV’s
deviation in tracking the path since disturbance information like wind and wave currents
is not directly measured. Nevertheless, after a significant deviation occurs, adjustments can
be made to reduce the deviation from the desired path and fulfill the purpose of tracking to
the desired path.
Sensors 2024, 24, 6589 21 of 26

Figure 18. Schematic diagram of USV path for target point tracking segment.

As shown in Figure 19, in order to visualize the effect of the USV tracking the desired
path, a vertical distance figure between the actual trajectory points and the desired path
is established. It can be observed from the figure that the vertical distance between the
USV and the desired path is consistently less than 0.5 m. Since the width of the ship is
0.8 m and the GPS module is located in the center of the USV, the distance from the GPS
module to either side of the USV is approximately 0.4 m. Thus, the deviation is within the
acceptable error range of 0.5 m. Initially, the vertical distance deviation caused by adjusting
the actual yaw angle was less than 0.2 m. Subsequently, due to interference factors, the
deviation increased. However, after adjustment, the vertical distance was significantly
reduced, indicating that the purpose of anti-interference was accomplished.

Figure 19. Vertical distance between the real-time position of the USV and the desired trajectory.

As shown in Figure 20, the desired yaw angle, ψd , as an input to the angular cascade
PID controller is based on the coordinates of the USV in conjunction with Equations (24)
and (25). The actual yaw angle, ψi , as the output and feedback to the angular cascade PID
controller is obtained directly from the IMU module. At the outset, the actual yaw angle
of the USV was 153.334°, while the desired heading angle was −130.515°. In combination
with the angle values of the coordinate system established in Section 4 and Figure 9, it
is determined that the USV needs to turn counterclockwise in order to turn through the
smallest angle to track the desired yaw angle, so the actual yaw angle first converges to
180° and then turns negative before converging to the desired yaw angle. As the motion
progresses, the angular cascade PID controller regulates the yaw angle by adjusting the
Sensors 2024, 24, 6589 22 of 26

rotational speed of the propellers, and the yaw angle deviation gradually approaches a
steady state, reflecting a excellent heading control effect and fulfilling the purpose of USV
heading maintenance.

Figure 20. Comparison of desired yaw angle and actual yaw angle of USV.

As shown in Figure 21, initially, the distance change is minimal due to the in-place
differential steering of the USV for adjusting the yaw angle, which is consistent with the
previous analysis. After adjusting the actual yaw angle to be consistent with the desired
yaw angle, the distance from the target point decreases rapidly, signifying that the USV is
moving swiftly toward the target point. Finally, the reduced speed for distance changes
indicates that the positional cascade PID controller regulates the speed of the USV and
prevents overshooting of the USV speed caused by excessive velocity.

Figure 21. Distance between the USV and the target point in the target point tracking segment.

5.2.2. Data Analysis of Dynamic Positioning Segment

As shown in Figure 22, latitude and longitude are converted to planar coordinates
following Mercator transformation for analysis. The positive direction of the Y-axis is
northward. Most of the paths shown in the figure are situated within a circle centered on
the target point with a radius of 1 m. At this juncture, the steering gear begins to operate in
conjunction with the propellers to adjust the actual position of the USV in the wind- and
wave-disturbed environment. As can be observed from the figure, after the USV enters into
the circle, it continuously adjusts near the target point. Since the GPS module is installed at
the center of the USV, the path shown in the figure is the outcome of the transformation of
the collected latitude and longitude information. Thus, the path showcases the adjustment
Sensors 2024, 24, 6589 23 of 26

process of the center of the USV. The length of the USV is 1 m and the width is 0.8 m.
Although the center of the USV cannot be constantly maintained at the target point, the
majority of the hull of the USV is within the circle and the distance from the target point is
small in comparison with the size of the boat. Therefore, it essentially accomplishes the
purpose of dynamic positioning.

Figure 22. Schematic diagram of USV path for dynamic positioning segment.

As shown in Figure 23, the relationship between yaw angles and rudder angles is
illustrated. Figure 23a shows yaw angles for the dynamic positioning segment. According
to Figure 22, it is evident that the USV enters different quadrants when adjusting its position,
resulting in significant variations in the azimuthal yaw angle, ψc , and the desired yaw
angle, ψd . In this segment, the USV controls rudder angles to distribute the thrust force.
Since the magnitude and direction of the thrust forces are the same on both sides, based
on Equation (14), it can be analyzed that the change in rudder angles of the USV does
not generate a moment. Hence, it does not affect the actual yaw angle, and the variation
in the actual yaw angle is insignificant. From the analysis of the flowchart presented in
Figure 15, it can be noted that if ψd > 0 and the value of ψi is within the range (ψc , ψd ), that
is, when the red line is between the yellow line and the blue line, ψel = ∆ψ in Figure 23b,
where the red line coincides with the green line. When the value of ψi is within the
range [ψd , 180] ∪ (−180, ψc ], that is, when the red line is outside the yellow and blue lines,
ψer = ∆ψ in Figure 23b, where the blue line coincides with the green line. Figure 23 also
displays the values of the rudder angle for ψd < 0, thereby verifying the feasibility of the
adaptive regulation principle.

(a) Desired yaw angle, actual yaw angle, and yaw angle deviation (b) Relation between rudder angles and yaw angle deviation

Figure 23. The analysis of rudder angles.

Sensors 2024, 24, 6589 24 of 26

As shown in Figure 24, in the dynamic positioning segment, the distance between
the actual position of the USV and the target position fluctuates up and down, suggesting
that the USV continuously adjusts its distance from the target point while remaining in the
vicinity of the target point. Although the amount of interference such as wind and waves
cannot be directly measured, it is inevitable that there are interference factors like wind
and waves when experiments are carried out in real water environments. The distance
gradually decreases from about 1 m to about 0.5 m. The distance between the USV and the
target fluctuates up and down yet shows an overall downward trend. This indicates that
the adaptive adjustment mechanism can reduce the deviation between the position of the
USV and the target in the interference environment, thereby proving the effectiveness of
the hardware and principle of the adaptive adjustment mechanism.

Figure 24. Distance between USV and target point in dynamic positioning segment.

6. Conclusions
The objective of this paper is to investigate the motion control system for a small
USV to rapidly track a target point and maintain it at the target point. To address the
problem of tracking and maintaining the heading, the heading coordinate system and
the ship coordinate system are established. Based on this, the value ranges of the angle
in the coordinate system are specified. The desired yaw angle value is calculated in real
time and compared with the actual yaw angle value to determine the working state of
the propeller for carrying out differential steering. Additionally, it is combined with the
angular cascade PID controller to regulate the speed of the propeller, enabling the USV to
track and maintain the desired heading. The motion process is a straight-line motion from
the starting point to the target point. Thus, path following is the tracking of the straight-line
path. Thanks to the real-time updating of the control system, as soon as the position of the
USV changes, the working state and rotational speed of the propeller are changed in real
time, and real-time corrections are made to the motion path, achieving good straight-line
path following results.
Upon the USV’s arrival at the target point, the existence of wind and waves makes
it challenging for the USV to stay at the target point. To address this issue, an adaptive
adjustment mechanism is designed to flexibly distribute the thrust force to counteract the
force exerted by the wind and waves on the USV, thereby reducing the offset of the USV’s
position. Finally, the effectiveness of the scheme presented in this paper is verified through
real ship experiments.

Author Contributions: Conceptualization, H.Z.; data curation, J.Z.; funding acquisition, H.Z.;
methodology, J.Z.; project administration, L.M.; software, J.Z.; supervision, L.M.; validation, Y.Y. and
K.L.; visualization, Z.F. and K.L.; writing—original draft, J.Z.; writing—review and editing, J.Z. All
authors have read and agreed to the published version of the manuscript.
Sensors 2024, 24, 6589 25 of 26

Funding: This work is supported by the Qilu University of Technology (Shandong Academy of
Sciences) Transformation Projects for Scientific and Technological Achievements (2023CGZH-02), the
Industry-University-Research Collaborative Innovation Fund Project (2020-CXY26), and the Jinan
Industrial Innovation Carrier Project (202333001).
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.
Data Availability Statement: The data that support the findings of this study are available from the
corresponding author upon reasonable request.
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Yang, X.; Yan, X.; Liu, W.; Ye, H.; Du, Z.; Zhong, W. An improved stanley guidance law for large curvature path following of
unmanned surface vehicle. Ocean Eng. 2022, 266, 112797. [CrossRef]
2. Liang, X.; Qu, X.; Wang, N.; Li, Y. Swarm velocity guidance based distributed finite-time coordinated path-following for uncertain
under-actuated autonomous surface vehicles. ISA Trans. 2021, 112, 271–280. [CrossRef] [PubMed]
3. Yang, X.; Shi, Y.; Liu, W.; Ye, H.; Zhong, W.; Xiang, Z. Global path planning algorithm based on double DQN for multi-tasks
amphibious unmanned surface vehicle. Ocean Eng. 2022, 266, 112809. [CrossRef]
4. Liu, Z.; Zhang, Y.; Yu, X.; Yuan, C. Unmanned surface vehicles: An overview of developments and challenges. Annu. Rev. Control
2016, 41, 71–93. [CrossRef]
5. Karimi, H.R.; Lu, Y. Guidance and control methodologies for marine vehicles: A survey. Control Eng. Pract. 2021, 111, 104785.
[CrossRef]
6. He, Z.; Fan, Y.; Wang, G.; Mu, D.; Song, B. Global finite-time PLOS-OCCs guidance and adaptive integral sliding mode path
following control for unmanned surface vehicles with ocean currents and input saturation. Ocean Eng. 2024, 299, 117227.
[CrossRef]
7. Mu, D.; Li, J.; Wang, G.; Fan, Y. Disturbance rejection control of adaptive integral LOS unmanned ship path following based on
fast wave inversion. Appl. Ocean Res. 2024, 144, 103907. [CrossRef]
8. He, Z.; Fan, Y.; Wang, G.; Mu, D. Cooperative trajectory tracking control of MUSVs with periodic relative threshold event-triggered
mechanism and safe distance. Ocean Eng. 2023, 269, 113541. [CrossRef]
9. Roberts, G. Trends in marine control systems. Annu. Rev. Control 2008, 32, 263–269. [CrossRef]
10. Gupta, D.K.; Dei, G.; Soni, A.K.; Jha, A.V.; Appasani, B.; Bizon, N.; Srinivasulu, A.; Nsengiyumva, P. Fractional order PID
controller for load frequency control in a deregulated hybrid power system using Aquila Optimization. Results Eng. 2024,
23, 102442. [CrossRef]
11. Wang, L.; Li, Y.; Liao, Y.; Pan, K.; Zhang, W. Adaptive heading control of unmanned wave glider with heading information fusion.
Control Eng. Pract. 2019, 85, 216–224. [CrossRef]
12. Wu, J.F.; Liu, L.L.; Zhong, X.H. MFA controller with tracking differentiator. Electr. Mach. Control 2007, 11, 547–550.
13. Bu, X.; Wang, Q.; Hou, Z.; Qian, W. Data driven control for a class of nonlinear systems with output saturation. ISA Trans. 2018,
81, 1–7. [CrossRef] [PubMed]
14. Li, Y.; Wang, L.; Liao, Y.; Jiang, Q.; Pan, K. Heading MFA control for unmanned surface vehicle with angular velocity guidance.
Appl. Ocean Res. 2018, 80, 57–65. [CrossRef]
15. Liao, Y.; Jiang, Q.; Du, T.; Jiang, W. Redefined Output Model-Free Adaptive Control Method and Unmanned Surface Vehicle
Heading Control. IEEE J. Ocean. Eng. 2020, 45, 714–723. [CrossRef]
16. Wang, N.; Sun, Z.; Yin, J.; Zou, Z.; Su, S.F. Fuzzy unknown observer-based robust adaptive path following control of underactuated
surface vehicles subject to multiple unknowns. Ocean Eng. 2019, 176, 57–64. [CrossRef]
17. Shen, Z.; Liu, Y.; Nie, Y.; Yu, H. Prescribed performance LOS guidance-based dynamic surface path following control of unmanned
sailboats. Ocean Eng. 2023, 284, 115182. [CrossRef]
18. Fossen, T.I.; Pettersen, K.Y.; Galeazzi, R. Line-of-Sight Path Following for Dubins Paths With Adaptive Sideslip Compensation of
Drift Forces. IEEE Trans. Control Syst. Technol. 2015, 23, 820–827. [CrossRef]
19. Fossen, T.I.; Grøvlen, Å. Nonlinear output feedback control of dynamically positioned ships using vectorial observer backstepping.
IEEE Trans. Control Syst. Technol. 1998, 6, 121–128. [CrossRef]
20. Mehrzadi, M.; Terriche, Y.; Su, C.L.; Othman, M.; Vasquez, J.C.; Guerrero, J.M. Review of Dynamic Positioning Control in
Maritime Microgrid Systems. Energies 2020, 13, 3188. [CrossRef]
21. Zhang, G.; Yao, M.; Zhang, W.; Zhang, W. Event-triggered distributed adaptive cooperative control for multiple dynamic
positioning ships with actuator faults. Ocean Eng. 2021, 242, 110124. [CrossRef]
Sensors 2024, 24, 6589 26 of 26

22. Du, J.; Hu, X.; Krstić, M.; Sun, Y. Dynamic positioning of ships with unknown parameters and disturbances. Control Eng. Pract.
2018, 76, 22–30. [CrossRef]
23. Fossen, T.I. Guidance and Control of Ocean Vehicles; Chichester: New York, NY, USA, 1994.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.