CN115951581A - High-speed unmanned ship path tracking control method based on improved EMPC - Google Patents

High-speed unmanned ship path tracking control method based on improved EMPC Download PDF

Info

Publication number
CN115951581A
CN115951581A CN202310018545.4A CN202310018545A CN115951581A CN 115951581 A CN115951581 A CN 115951581A CN 202310018545 A CN202310018545 A CN 202310018545A CN 115951581 A CN115951581 A CN 115951581A
Authority
CN
China
Prior art keywords
speed
unmanned ship
empc
unmanned
control method
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202310018545.4A
Other languages
Chinese (zh)
Inventor
戴晓强
李昂
黄鑫
曾庆军
王莹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Jiangsu University of Science and Technology
Original Assignee
Jiangsu University of Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Jiangsu University of Science and Technology filed Critical Jiangsu University of Science and Technology
Priority to CN202310018545.4A priority Critical patent/CN115951581A/en
Publication of CN115951581A publication Critical patent/CN115951581A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)

Abstract

The invention discloses a high-speed unmanned ship path tracking control method based on improved EMPC, which comprises the following steps: step 1: constructing a basic model and a coordinate system of the under-actuated water surface high-speed unmanned ship; step 2: in the guidance method, the radius of the LOS forward-looking circle is optimized by combining the current ship speed through a dynamic line-of-sight method, and an expected course angle is obtained on the basis of the optimized radius of the LOS forward-looking circle by combining a cross-side deviation rate; and 3, step 3: in the off-line state of the EMPC controller, performing optimal control law solving on each state partition in the off-line state through a white group optimization algorithm to obtain each state partition and linear control laws on corresponding partitions; and 4, step 4: and (4) in an online state of the EMPC controller, searching the linear control law on the corresponding subarea obtained in the step (3) by a reachable subarea searching method, and controlling the last expected path point on the unmanned ship to run and track so as to finish the tracking. The invention improves the control precision, reduces the lateral deviation and improves the self-adaptive capacity of the guidance parameters.

Description

High-speed unmanned ship path tracking control method based on improved EMPC
Technical Field
The invention relates to the technical field of high-speed unmanned ship control, in particular to a high-speed unmanned ship path tracking control method based on improved EMPC.
Background
The path tracking technology is used as an important guarantee for the unmanned ship to safely, autonomously, accurately and quickly complete various tasks, and the control target is to design a controller so that the unmanned ship can be accurately tracked and kept on a desired path independent of time in a tracking space. The problems of nonlinearity, model uncertainty, external interference time variation, parameter adaptability, controller robustness, tracking effect stability, navigation safety and the like in practical engineering application exist in the high-speed navigation process of the under-actuated unmanned ship, so that the path tracking controller with strong adaptive capacity, high robustness, strong anti-interference capacity and high real-time performance is designed, and the method plays an important role in the task adaptive capacity and the debugging safety of the high-speed unmanned ship.
The main control methods at present include a back-stepping method, sliding mode control, dynamic surface control, disturbance Observers (DO), neural networks, adaptive control, model predictive control, and the like. However, these methods have disadvantages in practical applications, and researchers at home and abroad have proposed various improvement schemes for these disadvantages. For example, an irrevocable term psi can appear in ship tracking control by Sun Z et al aiming at integral sliding mode c r c And further influencing the stability of the system, a method for practical proportional-integral sliding mode is provided, a heading angle and a position tracking error are additionally considered when a sliding mode surface is designed, and the problem is effectively solved. Chen Zhijuan aims at the influence of steering engine constraint and complex sea condition interference in the navigation process of the under-actuated unmanned ship, solves the rudder angle saturation problem by combining a model prediction control design controller, and achieves approximation to external interference through training of ship historical information by utilizing the training approximation characteristic of a radial basis function neural network, MPC is compensated, and the robustness of the system is improved. 5363 and combining LOS sight guidance algorithm and prediction function control aiming at the track control problem of the unmanned ship on the water surface, the Yang Tiantian provides a track control system based on LOS + PFC, and adopts a simulated annealing algorithm to solve the optimal control sequence of the prediction control algorithm, so as to further improve the accuracy and real-time performance of the track control.
The existing closest method is Chen Tianyuan, which is provided by a display model predictive control-based unmanned ship track control method research by a display model predictive control method and solves the problem of real-time requirement in the actual navigation of the unmanned ship by using the characteristics of offline calculation and online synthesis of the display model predictive control. According to the method, firstly, the unmanned ship track control is simplified into heading control by using a line of sight (LOS) method, and then a display model predictive control algorithm is applied to the problem of unmanned ship heading control, so that the control precision is ensured, the calculation speed is increased, and the real-time problem during high-speed navigation is solved to a certain extent.
According to the technical scheme closest to the text, although the calculation time is reduced to a certain extent, a certain control precision is lost in actual control, so that a proper algorithm is added to perform optimal solution on the objective function on each state partition in offline calculation to obtain a corresponding optimal control law, and the lost control precision is compensated. Meanwhile, the traditional LOS guidance method is weak in parameter adaptive capacity and prone to drift caused by environmental interference, so that linear guidance and curve guidance need to be improved respectively, lateral deviation is reduced, adaptive capacity during path switching is improved, and the USV can reach an expected path more quickly.
Disclosure of Invention
The invention provides a high-speed unmanned ship path tracking control method based on improved EMPC (empirical mode decomposition), which aims to solve the problems that the control precision is lost, the parameter self-adaptive capacity of a guidance method is weak, and drift angles are easily generated by environmental interference in the prior art.
The invention provides a high-speed unmanned ship path tracking control method based on improved EMPC, which comprises the following steps:
step 1: constructing a dynamic model, a kinematic model, a fixed coordinate system and an unmanned ship carrier coordinate system of the under-actuated water surface high-speed unmanned ship;
and 2, step: in the guidance method, the radius of the LOS forward-looking circle is optimized by combining the current ship speed through a dynamic line-of-sight method, an expected course angle is obtained on the basis of the optimized radius of the LOS forward-looking circle by combining a cross-side deviation rate, and a deviation angle is obtained according to an actual course angle;
and step 3: in the off-line state of the EMPC controller of the control method, carrying out optimal control law solving on each state partition in the off-line state through an Egret group optimization algorithm to obtain linear control laws on each state partition and the corresponding partition;
and 4, step 4: in the online state of the EMPC controller of the control method, the linear control laws on the corresponding subareas obtained in the step 3 are searched by a reachable subarea searching method, the unmanned ship is controlled to run by the searched linear control rate until the unmanned ship tracks the last expected path point, and the tracking control process is finished.
Further, in the step 1, the disturbance model in the dynamic model of the under-actuated surface high-speed unmanned ship comprises: the wind interference force model, the wave interference force model and the flow interference force model have the following specific formulas:
Figure BDA0004040895310000031
wherein [ u, v, r ]] T Representing the amount of unmanned boat speed, here m ii (1,2,3) is expressed as unmanned boat inertial hydrodynamic force, which is the force generated by the inertia of the surrounding water flow when the USV is accelerated, and is specifically expressed as
Figure BDA0004040895310000032
d 11 =-X u ,d 22 =-Y v ,d 33 =-N r Is the hydrodynamic damping coefficient; tau is X For longitudinal thrust, τ N For rotational forces, τ wX For longitudinal disturbing forces, tau wY For transverse disturbing forces, tau wN Is a gyroscopic disturbance force.
Further, the wind disturbance force model is as follows:
Figure BDA0004040895310000041
Figure BDA0004040895310000042
Figure BDA0004040895310000043
in the formula,
Figure BDA0004040895310000044
is the relative wind speed, gamma R =tan -1 (u R /v R ) As a relative velocity, C X 、C Y Is the thrust coefficient; c N Is a moment coefficient; rho w Is the air density in kg/m 3 ;A T 、A L Respectively as transverse and longitudinal projected areas; l is the total length of the unmanned boat and is m; v R Is the wind speed, in m/s.
Further, the wave interference force model is:
Figure BDA0004040895310000045
wherein rho is the density of seawater, chi is the encounter angle, m is the division number of the spectrum frequency of the sea wave, xb, yb and Nb are test coefficients,
Figure BDA0004040895310000046
is the wave surface equation of the sea wave>
Figure BDA0004040895310000047
Is the wavelength.
Further, the flow disturbance force model is:
Figure BDA0004040895310000048
in the formula, F Hr =-C(v r )v r -D(v r )v r Acting force of fluid after disturbance of ocean current, v r =[u+u c ,v+v c ,r] T The projection of the relative speed of the unmanned surface vehicle movement to the water flow is obtained; f H And the force produced by the relative motion of the fluids is = C (V) V-D (V) V.
Further, in the step 2, the formula for optimizing the LOS forward-looking circle radius according to the current ship speed is as follows:
R K =e L +e -λU k L
in the formula, R K LOS front view circle radius; e.g. of the type t The vertical distance from the unmanned surface vehicle to the expected route at the current moment is obtained; u is the current navigational speed;
the specific formula for obtaining the expected heading angle by combining the lateral deviation rate on the basis of the optimized LOS front view circle radius is as follows:
Figure BDA0004040895310000051
in the formula,
Figure BDA0004040895310000052
Figure BDA0004040895310000053
further, the discrimination conditions in the aigret group optimization algorithm are as follows:
Figure BDA0004040895310000054
the invention has the beneficial effects that:
aiming at the problems of large parameter calculation amount and low solving efficiency of the traditional model predictive control in the high-speed unmanned ship path tracking process, the method adopts a method for displaying model predictive control, and simultaneously introduces an Egret group optimization algorithm in an off-line calculation part to improve the solving of the optimal control law of each state partition. The advantages of an ESOA algorithm and an EMPC controller are combined, the aim function is optimally solved by utilizing the advantages of low requirements of the Egret group optimization algorithm on parameters, initial values, the aim function and the like, the advantages of a distribution optimization strategy, good overall convergence, rapid convergence and the like, and the control accuracy of the path tracking of the high-speed unmanned ship is integrally improved. Meanwhile, the improved LOS guidance method comprises the compensation of drift angle generated by interference, so that the lateral deviation is reduced to a certain extent, and the self-adaptive capacity of the guidance parameters is improved.
Drawings
The features and advantages of the present invention will be more clearly understood by reference to the accompanying drawings, which are schematic and are not to be understood as limiting the invention in any way, and in which:
FIG. 1 is a schematic view of a line-of-sight guidance according to an embodiment of the present invention;
FIG. 2 is a block diagram of an Egret group optimization algorithm in an embodiment of the present invention;
FIG. 3 is a comparison graph of the convergence curves of the optimization algorithm in an embodiment of the present invention.
FIG. 4 is a flow chart of a reachable partition algorithm in an embodiment of the present invention;
FIG. 5 is a block diagram of a control system in an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the invention provides a high-speed unmanned ship path tracking control method based on improved EMPC, which comprises the following steps:
step 1: and (3) constructing a kinematics and dynamics mathematical model of the under-actuated water surface high-speed unmanned ship. A Fossen ship dynamics mathematical model is adopted to simplify a ship six-degree-of-freedom model into an unmanned ship motion model with three degrees of freedom, namely, surging, swaying and yawing.
1) The water surface high-speed unmanned ship kinematics model comprises the following steps:
Figure BDA0004040895310000071
2) The dynamic model of the water surface high-speed unmanned ship is as follows:
Figure BDA0004040895310000072
wherein [ x, y, ψ] T Represents the position vector of the unmanned ship, [ u, v, r] T Representing the amount of unmanned boat speed, here m ii (1,2,3) is expressed as unmanned boat inertial hydrodynamic force, which is the force generated by the inertia of the surrounding water flow when the USV is accelerated, and is specifically expressed as
Figure BDA0004040895310000073
d 11 =-X u ,d 22 =-Y v ,d 33 =-N r Is the hydrodynamic damping coefficient; tau. x And is longitudinal thrust, τ N For rotational forces, τ wX For longitudinal disturbing forces, tau wY For transverse disturbing forces, tau wN Is a gyroscopic disturbance force. In the above formula, X u
Figure BDA0004040895310000074
Y v
Figure BDA0004040895310000075
Y r
Figure BDA0004040895310000076
N v
Figure BDA0004040895310000077
N r
Figure BDA0004040895310000078
I z Is the hydrodynamic coefficient.
The unmanned ship in the embodiment can only generate longitudinal thrust tau x And a turning force τ N In order to make the model more accurate, an external interference model is additionally added, and is defined as follows:
Figure BDA0004040895310000079
in the formula,
Figure BDA0004040895310000081
is a wind disturbance force model>
Figure BDA0004040895310000082
Is a wave disturbance force model>
Figure BDA0004040895310000083
Is a flow disturbance force model.
Wind speed generally comprises a slowly varying component (mean wind speed) and a high frequency component (gust), and the resultant forces and moments on a surface drone are generally dependent on the relative wind speed V R And relative velocity γ R Depending on:
Figure BDA0004040895310000084
(5)
γ R =tan -1 (u R /v R )
V R the resultant speed in the x direction and the y direction is considered, the relative wind speed of the unmanned surface vehicle in the sailing state is considered, and the wind speed is decomposed along a follow-up coordinate system:
(6)
u R =V w cos(ψ R )-u+u c
v R =V w sin(ψ R )-v+v c
in the formula, /) R =ψ-ψ w The relative angle between the wind direction and the ship heading is adopted, and for most unmanned water surface boats, gust cannot be compensated through a control system. But calculating slowly changing wind power through wind speed and wind direction, and feeding the slowly changing wind power to the controller in front to act on the wind power on the shipThe torque formula is as follows:
Figure BDA0004040895310000085
Figure BDA0004040895310000086
Figure BDA0004040895310000087
in the formula, C X 、C Y Is the thrust coefficient; c N Is a moment coefficient; rho w Is the air density in kg/m 3 ;A T 、A L Respectively as transverse and longitudinal projected areas; l is the total length of the unmanned boat and is m; v R Is the wind speed, in m/s.
For waves at sea, the waves are generally described by linear superposition, in order to forecast the interference force borne by a water surface unmanned ship in irregular waves, a wave spectrum counted according to a large amount of wave observation data is generally adopted to describe the waves in a wave tolerance theory of the water surface unmanned ship, the commonly used wave spectrum comprises a PM spectrum, a single-parameter spectrum and a double-parameter spectrum, the ITTC double-parameter wave spectrum recommended by an international ship model pool conference is adopted to forecast the irregular waves, and the formula is as follows:
Figure BDA0004040895310000091
in the formula, ζ W/3 Three-one wave height, omega wave frequency, T 1 Is the average wave period.
According to the wave spectrum, the wave surface equation of the wave can be obtained:
Figure BDA0004040895310000092
the wavelength is as follows:
Figure BDA0004040895310000093
finally, according to the formula of the acting force of the waves on the hull, the formula for calculating the acting force of the sea waves on the unmanned surface vehicle can be obtained as follows:
Figure BDA0004040895310000094
in the formula, rho is the density of seawater, chi is an encounter angle, m is the division number of the wave spectrum frequency, xb, yb and Nb are test coefficients, and can be estimated according to a regression formula obtained by the test of J.W English and the like:
Figure BDA0004040895310000101
recording the flow velocity of ocean current as V c And the flow direction is beta, and the flow velocity is decomposed along a sea surface fixed coordinate system:
Figure BDA0004040895310000102
the coordinate conversion formula can be used for obtaining:
Figure BDA0004040895310000104
substituting the formulas 3-22 into the above formula to obtain:
Figure BDA0004040895310000105
the components of the movement of the unmanned surface vehicle to the ground speed on the ship body coordinate system are u and v, and the projection of the movement of the unmanned surface vehicle to the relative speed of the water flow is as follows:
(16)
v r =[u+u c ,v+v c ,r] T
for the water surface unmanned ship dynamic model, the acting forces generated by the relative motion of the fluid are as follows:
(17)
F H =-C(v)v-D(v)v
by replacing the actual speed with the relative speed, the fluid acting force after the ocean current disturbance can be obtained:
(18)
F Hr =-C(v r )v r -D(v r )v r
after difference, ocean current acting force can be obtained:
Figure BDA0004040895310000111
step 2: in the guidance method, the radius of the LOS foresight park is optimized by combining the current ship speed through a dynamic line-of-sight method, an expected course angle is obtained on the basis of the optimized radius of the L0S foresight park by combining a lateral deviation rate, and a deviation angle is obtained according to an actual course angle. Suppose that the position of the unmanned ship at the current moment is a point p t (x t ,y t ) The expected circle center of the tracking point is a point p k (x k ,y k ) As shown in fig. 1.
Obtaining a target angle according to the current position of the unmanned ship and the expected circle center position:
Figure BDA0004040895310000112
the expected path points are obtained according to the circle tracking direction as follows:
Figure BDA0004040895310000113
Figure BDA0004040895310000114
then the desired heading angle may be obtained:
Figure BDA0004040895310000115
under the complex sea condition environment, when the unmanned ship tracks the circular path, a drift angle which causes interference to the navigation direction is generated, and in order to eliminate the influence of the drift angle, and according to the lateral deviation and the lateral deviation change law, an S-plane is designed to control to obtain a radian deviation correction guidance law so as to guide the unmanned ship to track the circle.
The lateral deviation of the unmanned boat from the expected path is as follows:
Figure BDA0004040895310000121
further, the lateral deviation change rate can be:
Figure BDA0004040895310000122
the expected course included angle of the unmanned ship is as follows:
Figure BDA0004040895310000123
the expected course of the unmanned ship after rectification can be obtained according to the formula as follows:
Figure BDA0004040895310000124
in the formula,
Figure BDA0004040895310000125
therefore, again based on the actual course angle
Figure BDA0004040895310000126
The difference with the desired heading angle results in a deflection angle pick>
Figure BDA0004040895310000127
Figure BDA0004040895310000128
And updating the path segment. In the selection of path P n+1 Then, whether the unmanned ship is P or not is judged n If the circle center is the circle center and R is the radius, the next path P is tracked n+1 . Let the current position of the unmanned ship be (x) n (t),y n (t)) satisfies:
Figure BDA0004040895310000129
will select (x) n+1 (t),y n+1 (t)) as the end point of the next path. Dynamically adjusting the radius of a forward-looking circle of the guidance law of the unmanned surface vehicle according to the vertical distance between the unmanned surface vehicle and the expected path and the current speed of the unmanned surface vehicle:
(23)
R K =e L +e -λU k L
wherein e is L The vertical distance between the unmanned surface vehicle and the expected course at the current moment is obtained.
The predictive control model is subjected to linearization processing, and the expected course angle is processed through a given expected state and an LOS algorithm
Figure BDA0004040895310000131
Deflection angle based on the actual course angle of the current state>
Figure BDA0004040895310000132
To track the desired path. Equation (32) represents the reference system equation, i.e., without considering the interference situationThe following reference trajectories:
Figure BDA0004040895310000133
the function is subjected to a first order Taylor expansion at an arbitrary reference point (xR, uR) to obtain the formula (33)
Figure BDA0004040895310000134
Subtracting (32) from formula (33) to obtain
Figure BDA0004040895310000135
Wherein,
Figure BDA0004040895310000136
A. b is a Jacobian matrix.
The new prediction model, i.e., equation (34), is discretized. There are many discretization methods, such as the longge-kutta method, where the forward euler method is used to obtain equation (35):
Figure BDA0004040895310000137
wherein, T is sampling time, I is a unit matrix, and the combination formula can be obtained as follows:
Figure BDA0004040895310000138
Figure BDA0004040895310000141
wherein
Figure BDA0004040895310000142
P is an identity matrix.
And transforming the formula to be simplified into a state space model in a control increment form:
Figure BDA0004040895310000143
Figure BDA0004040895310000144
wherein
Figure BDA0004040895310000145
Figure BDA0004040895310000146
And step 3: in the off-line state of the EMPC controller of the control method, under the setting of constraint conditions, carrying out optimal control law solving on each state partition in the off-line state through an Egret group optimization algorithm to obtain linear control laws on each state partition and the corresponding partition.
And (3) constraint condition setting: as the propeller and the steering engine of the unmanned ship are influenced by mechanical properties, the motion performance and the speed are limited, and the unmanned ship is easy to saturate during high-speed navigation. Therefore, the control quantity limit, the control increment and the output quantity are restricted in the k time and the prediction time domain.
Optimal solution of the objective function: taking into account the rate of the target and the loss of the control quantity energy of the system input. And constructing an objective function by using the state quantity deviation, the control quantity and the control increment of the system, wherein the objective function is as follows:
(32)
J=(Y-Yref) T Q(Y-Yref)+ΔU T RΔU
where Yref is the desired value, Δ U is the control increment, and Q and R are the weight matrices (constantly adjusted as needed for control).
The aigrette swarm optimization algorithm consists of three main components: sitting, etc. strategies, radical strategies and discrimination conditions. The algorithm block diagram is shown in fig. 2.
Each group of the white aigres can be composed of n white aigres, each white aigres comprises three white aigres, wherein the white aigres A implement sitting and other strategies, and the white aigres B and C respectively adopt random walking and surrounding mechanisms in an aggressive strategy.
1) Sit, etc. strategy (aigret a). The observation equation for the ith Egret A can be described as
Figure BDA0004040895310000151
True fitness y obtained by each iteration i The pseudo-gradient g of the weight in the observation equation can be found i The updated location of egru a is then represented as follows:
X a,i =X i +exp(-t/(0.1*t max ))*0.1*hop*g i (41)
where t is the current iteration number, t max For the maximum number of iterations, hop is the feasible domain range of the argument.
2) Aggressive strategies (Egret B, egret C). The aigrette B is randomly walked, and the position updating mode is as follows:
X b,i =X i +tan(r b,i )*hop/(1+t) (42 )
r b,i is a random number between (-pi/2, pi/2).
Egret C is a bounding strategy, and the positions are updated in the following way:
D h =X ibest -X i (43)
D g =x gbest -X i (44 )
X c,i =(1-r i -r g )*X i +r h *D h +r g *D g (45 )
wherein X ibest And X gbest Respectively an Egret team optimum and an Egret population optimum, r h And r g Are all [0,1]A random number in between.
3) And (5) judging the condition. After each aigrette of the aigrette calculates the updated position, the updated positions of the aigrette teams are determined together, and the form is as follows:
X Si= [X a,i X b,i X c,i ] (46)
y s,i =[y a,i y b,i y c,i ] (47)
c i =argmin(y s,i ) (48)
Figure BDA0004040895310000161
the positions and the fitness of the three updated aigres are compared with the fitness of the last iteration by the aigres, and if the updated position of one aigret is better than the position of the last iteration, the update is adopted; if the update location of each of the egrts is worse than the previous one, there is a 33% probability that the solution with the best update location will be adopted.
Fig. 3 is a convergence curve comparison graph of the aigrette swarm optimization algorithm and the particle swarm optimization algorithm, wherein the image on the left half is a target function search space, and the image on the right half is a convergence curve of the two optimization algorithms.
Solving the optimal control increment sequence by utilizing an aigret group optimization algorithm to obtain an expression (50):
ΔU * =[Δu(t|t) * Δu(t+1|t) * … Δu(t+Nc-1|t) * ] (50)
and 4, step 4: in the online state of the EMPC controller of the control method, online searching is carried out by adopting a reachable partition searching method according to the actual state quantity output by the high-speed unmanned ship sensor, the state partition is judged, the corresponding optimal control law is searched, and the control component is applied to the unmanned ship.
The reachable partition search algorithm flowchart is shown in fig. 4, and includes the following specific steps:
1) For the initial state x 0 Positioning the corresponding sub-region n by a sequential searching method;
2) Obtaining a corresponding control quantity u according to the sub-region n;
3) Applying the control quantity to the system, and detecting the system state quantity at the next moment;
4) And searching the state partition where the state quantity is located according to the state quantity of the system and the reachable partition.
5) Repeating (2), (3) and (4) until the table lookup is finished.
And applying the optimal control component formula obtained by table lookup to the high-speed unmanned ship.
u(t) * =Δu(t|t) * +u(t-1) (51)
The whole control system is shown in a block diagram in fig. 5, and a prediction control method is improved by adopting a LOS-ESOA-EMPC control method. The method comprises the steps of establishing a dynamic model of the under-actuated unmanned ship as a state space model of predictive control, carrying out linearization, discretization and other processing on the state space model to obtain a control incremental equation in an error form as a system model, setting an expected path curve, uniformly dividing an expected path into a plurality of expected points, obtaining position information of the expected points, and calculating an expected course angle by adopting an improved LOS guidance algorithm according to the current state of the unmanned ship. The method comprises the steps that an error between an expected course angle and an actual course angle serves as input of an EMPC controller, during off-line calculation, under the setting of an initial state and constraint conditions, an Egret group optimization algorithm (ESOA) is adopted to conduct objective function optimization solving, linear control laws on each state partition area and corresponding partitions are obtained, during on-line calculation, the state partitions are judged according to actual state quantity output by a high-speed unmanned ship sensor, corresponding control laws are searched, and control components are applied to the unmanned ship. And (3) repeating the steps 1-3 according to the planned expected path, continuously feeding back the actual value of the high-speed unmanned ship measuring sensor to the EMPC controller at the next moment, and continuously searching on line until the high-speed unmanned ship tracks to the last expected point.
Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations fall within the scope defined by the appended claims.

Claims (7)

1. A high-speed unmanned ship path tracking control method based on improved EMPC is characterized by comprising the following steps:
step 1: constructing a dynamic model, a kinematic model, a fixed coordinate system and an unmanned ship carrier coordinate system of the under-actuated water surface high-speed unmanned ship;
step 2: in the guidance method, the radius of the LOS forward-looking circle is optimized by combining the current ship speed through a dynamic line-of-sight method, an expected course angle is obtained on the basis of the optimized radius of the LOS forward-looking circle by combining a cross-side deviation rate, and a deviation angle is obtained according to an actual course angle;
and step 3: in the off-line state of the EMPC controller of the control method, carrying out optimal control law solving on each state partition in the off-line state through a Egret group optimization algorithm to obtain each state partition and a linear control law on a corresponding partition;
and 4, step 4: in the online state of the EMPC controller of the control method, the linear control laws on the corresponding subareas obtained in the step 3 are searched through a reachable subarea searching method, the unmanned ship is controlled to run by the searched linear control rate until the unmanned ship tracks the last expected path point, and the tracking control process is completed.
2. The EMPC-based high-speed unmanned ship path tracking control method of claim 1, wherein in the step 1, the interference model in the dynamic model of the under-actuated surface high-speed unmanned ship comprises: the wind interference force model, the wave interference force model and the flow interference force model have the following specific formulas:
Figure FDA0004040895300000011
wherein [ u, v, r] T Representing the amount of unmanned boat speed, here m ii (1,2,3) is expressed as unmanned boat inertial hydrodynamic force, which is the force generated by the inertia of the surrounding water flow when the USV is accelerated, and is specifically expressed as
Figure FDA0004040895300000025
d 11 =-X u ,d 22 =-Y v ,d 33 =-N r Is the hydrodynamic damping coefficient; tau is X For longitudinal thrust, τ N For rotational forces, τ wX For longitudinal disturbing forces, tau wY For transverse disturbing forces, tau wN Is a gyroscopic disturbance force.
3. The EMPC-based high-speed unmanned ship path tracking control method of claim 2, wherein the wind disturbance force model is:
Figure FDA0004040895300000021
Figure FDA0004040895300000022
Figure FDA0004040895300000023
in the formula,
Figure FDA0004040895300000024
is the relative wind speed, gamma R =tan -1 (u R /v R ) As a relative velocity, C X 、C Y Is the thrust coefficient; c N Is a moment coefficient; rho w Is the air density; a. The T 、A L Respectively are transverse and longitudinal projection areas; l is the total length of the unmanned boat; v R Is the wind speed.
4. The EMPC-based high-speed unmanned ship path tracking control method of claim 2, wherein the wave disturbance force model is:
Figure FDA0004040895300000031
wherein rho is the density of seawater, chi is the encounter angle, m is the division number of the spectrum frequency of the sea wave, xb, yb and Nb are test coefficients,
Figure FDA0004040895300000032
in the wave surface equation of sea waves>
Figure FDA0004040895300000033
Is the wavelength.
5. The EMPC-based high-speed unmanned ship path tracking control method of claim 2, wherein the flow disturbance force model is:
Figure FDA0004040895300000034
in the formula, F Hr =-C(v r )v r -D(v r )v r Acting force of fluid after disturbance of ocean current, v r =[u+u c ,v+v c ,r] T The projection of the relative speed of the unmanned surface vehicle movement to the water flow is obtained; f H And the force produced by the relative motion of the fluids is = C (v) v-D (v) v.
6. The method for high-speed unmanned ship path tracking control based on EMPC as claimed in claim 1, wherein in step 2, the formula for optimizing LOS forward-looking circle radius by current ship speed is as follows:
R K =e L +e -λU k L
in the formula, R K LOS front view circle radius; e.g. of the type t The vertical distance from the unmanned surface vehicle to the expected course at the current moment is determined; u is the current navigational speed;
the specific formula for obtaining the expected heading angle by combining the lateral deviation rate on the basis of the optimized LOS front view circle radius is as follows:
Figure FDA0004040895300000041
in the formula,
Figure FDA0004040895300000042
Figure FDA0004040895300000043
7. the EMPC-based high-speed unmanned ship path tracking control method of claim 1, wherein the criteria in the Egret group optimization algorithm are:
Figure FDA0004040895300000044
CN202310018545.4A 2023-01-06 2023-01-06 High-speed unmanned ship path tracking control method based on improved EMPC Pending CN115951581A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310018545.4A CN115951581A (en) 2023-01-06 2023-01-06 High-speed unmanned ship path tracking control method based on improved EMPC

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310018545.4A CN115951581A (en) 2023-01-06 2023-01-06 High-speed unmanned ship path tracking control method based on improved EMPC

Publications (1)

Publication Number Publication Date
CN115951581A true CN115951581A (en) 2023-04-11

Family

ID=87289300

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310018545.4A Pending CN115951581A (en) 2023-01-06 2023-01-06 High-speed unmanned ship path tracking control method based on improved EMPC

Country Status (1)

Country Link
CN (1) CN115951581A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116164753A (en) * 2023-04-18 2023-05-26 徐州徐工重型车辆有限公司 Mine unmanned vehicle path navigation method and device, computer equipment and storage medium
CN118102325A (en) * 2024-04-19 2024-05-28 华东交通大学 Three-dimensional directed sensor network coverage control method

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116164753A (en) * 2023-04-18 2023-05-26 徐州徐工重型车辆有限公司 Mine unmanned vehicle path navigation method and device, computer equipment and storage medium
CN116164753B (en) * 2023-04-18 2023-08-08 徐州徐工重型车辆有限公司 Mine unmanned vehicle path navigation method and device, computer equipment and storage medium
CN118102325A (en) * 2024-04-19 2024-05-28 华东交通大学 Three-dimensional directed sensor network coverage control method

Similar Documents

Publication Publication Date Title
CN108427414B (en) Horizontal plane self-adaptive trajectory tracking control method for autonomous underwater vehicle
CN106444794B (en) A kind of printenv drive lacking UUV vertical plane path trace sliding-mode control
Wang et al. Line-of-sight guidance law for path following of amphibious hovercrafts with big and time-varying sideslip compensation
CN115951581A (en) High-speed unmanned ship path tracking control method based on improved EMPC
Saoud et al. Routing and course control of an autonomous sailboat
CN109656142B (en) Cascade structure model-free self-adaptive guidance method for unmanned ship
CN110609556A (en) Multi-unmanned-boat cooperative control method based on LOS navigation method
Yin et al. Predictive trajectory tracking control of autonomous underwater vehicles based on variable fuzzy predictor
Shen et al. Path-following control of underactuated ships using actor-critic reinforcement learning with MLP neural networks
CN109916419A (en) A kind of hybrid genetic algorithm unmanned boat real-time route planing method of object-oriented
CN113359737A (en) Ship formation self-adaptive event trigger control method considering formation expansion
Hu et al. Trajectory tracking and re-planning with model predictive control of autonomous underwater vehicles
CN112462777B (en) Ship formation path active coordination system and method considering maneuverability difference
Xu et al. Waypoint-following for a marine surface ship model based on vector field guidance law
CN115014355A (en) Fixed-point return regulation and control method and device for catamaran unmanned ship
Fan et al. Path-Following Control of Unmanned Underwater Vehicle Based on an Improved TD3 Deep Reinforcement Learning
CN113296505B (en) Unmanned ship multi-mode path tracking control method based on speed change LOS
CN114967702A (en) Unmanned ship control system and path tracking method
CN113960998A (en) Unmanned ship fuzzy prediction control system and method
CN114035567B (en) Unmanned surface vehicle navigation control system
Peng et al. Safety-Certificated Line-of-Sight Guidance of Unmanned Surface Vehicles for Straight-Line Following in a Constrained Water Region Subject to Ocean Currents
Bibuli et al. An advanced guidance & control system for an unmanned vessel with azimuthal thrusters
Zheng et al. Nonlinear model predictive path following for an unmanned surface vehicle
Hu et al. Collision avoidance of USV by model predictive control-aided deep reinforcement learning
CN117519136B (en) Unmanned ship path tracking method considering large curvature turning

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination