Unmanned ship trajectory tracking control method and device based on improved model predictive control

Abstract

The application discloses a kind of based on improved model predictive control unmanned ship trajectory tracking control method and device.The method includes, H-PINN method is introduced into unmanned ship hydrodynamics model, obtains the simplified hydrodynamics model of unmanned ship;Based on the prediction model of the trajectory tracking of unmanned ship obtained in simplified hydrodynamics model, the expected trajectory and control quantity of model predictive control MPC are designed using differential flatness theory, and stability analysis is designed according to Lyapunov function.The method improves model predictive control using differential flatness theory, and Lyapunov function optimization analysis is combined, improves trajectory tracking precision, reduces the change amplitude of control quantity, so that the speed change in navigation process is more smooth.

Description

Technical Field

[0001] This invention relates to the field of unmanned vessel trajectory tracking and control technology, and in particular to an unmanned vessel trajectory tracking and control method and apparatus based on improved model predictive control. Background Technology

[0002] Unmanned surface vessels (USVs) are important tools in marine resource exploration and development, possessing the ability to perform diverse tasks in varied and complex aquatic environments. They can significantly reduce manpower costs, improve mission safety, and demonstrate strong resilience under extreme conditions. However, their maneuverability is affected by various environmental factors and is crucial to navigation safety. Therefore, accurately predicting their maneuverability and pre-designing motion control systems are essential.

[0003] Most modern unmanned surface vessels (USVs) are typically equipped with a single-propeller, single-rudder, or twin-propeller propulsion architecture, rarely featuring lateral propulsion devices with lateral thrust output, making them a typical underactuated system. Considering that USVs are inevitably subject to environmental disturbances from hydrodynamics, wind, waves, and currents, and are also dynamic and highly nonlinear systems, the underactuation, dynamic uncertainties, and time-varying disturbances of USVs make current trajectory tracking research based on dynamic models extremely challenging. Summary of the Invention

[0004] Model Predictive Control (MPC) not only excels at handling various constraint problems, but its application in unmanned surface vessel (USV) tracking control significantly improves the accuracy of tracking results. Therefore, in-depth research into more efficient and robust USV trajectory tracking control technologies has significant contemporary value. This invention provides an USV trajectory tracking control method and apparatus based on improved model predictive control. It utilizes differential flatness theory to improve model predictive control and combines Lyapunov function optimization analysis to enhance trajectory tracking accuracy, reduce the amplitude of control variable changes, and make speed changes during navigation smoother.

[0005] In a first aspect, embodiments of the present invention provide an unmanned surface vessel trajectory tracking control method based on improved model predictive control, comprising:

[0006] By introducing the H-PINN method into the hydrodynamic model of the unmanned vessel, a simplified hydrodynamic model of the unmanned vessel is obtained.

[0007] Based on the simplified hydrodynamic model, a prediction model for unmanned vessel trajectory tracking is obtained. The model is designed using differential flatness theory to predict the desired trajectory and control variable of the MPC. Stability analysis is designed based on the Lyapunov function.

[0008] Secondly, embodiments of the present invention provide an unmanned vessel trajectory tracking control device based on improved model predictive control, comprising:

[0009] The model building module is used to introduce the H-PINN method into the hydrodynamic model of the unmanned vessel to obtain a simplified hydrodynamic model of the unmanned vessel.

[0010] The model predictive control optimization module is used to obtain a predictive model for unmanned vessel trajectory tracking based on the simplified hydrodynamic model, design the desired trajectory and control quantity of the model predictive control (MPC) using differential flatness theory, and design stability analysis based on the Lyapunov function.

[0011] Thirdly, embodiments of the present invention provide a computer storage medium storing computer-executable instructions, which, when executed by a processor, implement the aforementioned unmanned vessel trajectory tracking control method based on improved model predictive control.

[0012] Fourthly, embodiments of this disclosure provide a server, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the above-described unmanned vessel trajectory tracking control method based on improved model predictive control.

[0013] The beneficial effects of the above-described technical solutions provided in the embodiments of the present invention include at least the following:

[0014] (1) The unmanned vessel trajectory tracking control method based on improved model predictive control provided in this embodiment of the invention uses differential flatness theory to design the desired trajectory and control quantity that conforms to the unmanned vessel system model, and applies its variables to model predictive control to improve trajectory tracking accuracy. In addition, in response to the problem of control quantity overshoot caused by large deviation between the initial position and the desired trajectory of the unmanned vessel trajectory tracking, the Lyapunov function optimization analysis is combined to guide the tracking of the desired trajectory, effectively solving the problem of large deviation between the initial and desired state quantities, reducing the amplitude of control quantity change, and making the speed change during navigation smoother.

[0015] (2) The unmanned vessel trajectory tracking control method based on improved model predictive control provided in this embodiment of the invention introduces the H-PINN method into the hydrodynamic model of the unmanned vessel and uses a physics-driven deep learning method instead of a data-driven deep learning method, resulting in a more accurate unmanned vessel dynamic model.

[0016] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings.

[0017] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0018] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0019] Figure 1 This is a flowchart of the unmanned vessel trajectory tracking control method based on improved model predictive control in an embodiment of the present invention;

[0020] Figure 2 This is a model of the surface motion of an unmanned vessel in an embodiment of the present invention;

[0021] Figure 3 for Figure 1 The detailed implementation flowchart of step S11 is shown below;

[0022] Figure 4 This is a schematic diagram of the model predictive control principle in an embodiment of the present invention;

[0023] Figure 5 This is an example diagram showing the prediction results of the unmanned surface vessel's motion attitude using the H-PINN method in an embodiment of the present invention;

[0024] Figure 6 This is a schematic diagram of the unmanned vessel trajectory tracking control device based on improved model predictive control in an embodiment of the present invention. Detailed Implementation

[0025] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0026] It should be understood that the terminology used in this invention is merely for describing particular embodiments and is not intended to limit the invention. Furthermore, with respect to numerical ranges in this invention, it should be understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. Every smaller range between any stated value or intermediate value within a stated range, and any other stated value or intermediate value within said range, is also included in this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0027] Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. While only preferred methods and materials have been described herein, any methods and materials similar or equivalent to those described herein may be used in the implementation or testing of this invention. All references to this specification are incorporated by way of citation to disclose and describe methods and / or materials associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0028] An unmanned surface vessel (USV) trajectory tracking control method and device based on improved model predictive control utilizes differential flatness theory to improve model predictive control and combines Lyapunov function optimization analysis to improve trajectory tracking accuracy, reduce the amplitude of control variable changes, and make speed changes during navigation smoother.

[0029] Example

[0030] This invention provides a method for unmanned surface vessel trajectory tracking control based on improved model predictive control, the process of which is as follows: Figure 1 As shown, it includes the following steps:

[0031] Step S11: Introduce the H-PINN method into the hydrodynamic model of the unmanned vessel to obtain a simplified hydrodynamic model of the unmanned vessel.

[0032] First, the establishment of the hydrodynamic model of the unmanned vessel is introduced, including the following steps:

[0033] 1. Construct an overall hydrodynamic model of the unmanned vessel in the three-degree-of-freedom dynamic model.

[0034] See Figure 2 The figure shows the surface motion model of the unmanned surface vessel (USV), where the ξ-E-η coordinate system is an inertial coordinate system, (x, y, ψ) represents the position of the USV, ψ represents the heading angle, and x and y are the values ​​on the ξ and η axes, respectively; x b -O b -y b In body coordinate system, u represents longitudinal velocity, v represents lateral velocity, and r represents turning angular velocity; F u T represents the resultant force in its longitudinal direction. r This refers to the steering torque caused by the speed difference between the two propellers in the propulsion structure.

[0035] Constructing an integral hydrodynamic model of the unmanned vessel in a three-degree-of-freedom dynamic model:

[0036]

[0037] In equation (1), the symbol J(η) represents the rotation matrix; the symbol M represents the mass matrix, satisfying M = M RB +MA The mass of a component can be decomposed into two sources: one from the rigid body and the other from hydrodynamic forces. The symbol C(v) represents the Coriolis centripetal force matrix, which can also be decomposed into two sources: one from the rigid body and the other from hydrodynamic forces. The symbol D(v) represents the damping matrix, which can also be decomposed into two sources: linear damping and nonlinear damping. In this embodiment, the dot (·) above the parameter represents the first derivative with respect to time. For example, adding a dot above a parameter representing a certain velocity indicates the acceleration corresponding to that velocity.

[0038] The specific mathematical descriptions of the above four are as follows:

[0039]

[0040] In equation (2), m represents the mass of the unmanned vessel hull, and I ZZ The moment of inertia of the bow roll is represented by m. (·) and d (·) These are all intermediate parameters, X (·) Y (·) and N (·) All belong to the hydrodynamic coefficients defined by SNAME, specifically X (·) This refers to longitudinal force, Y (·) It is the sway force, N (·) This refers to the partial derivative of the torque with respect to the variable in the subscript of the symbol. Under the premise of low-speed navigation, the nonlinear damping term can be excluded from consideration; under the premise of high-speed navigation, the nonlinear damping must be included in the consideration. For widely used ordinary unmanned vessels, when constructing the corresponding mathematical model, it is only necessary to introduce Taylor's formula to expand to the second-order terms to meet the accuracy requirements of the model.

[0041] This embodiment is for an unmanned surface vessel using a dual-propeller propulsion structure. The vector description of the force / torque exerted by the propulsion device on the hull can be expressed as:

[0042] τ=[F u 0 T r ] T (3)

[0043] In equation (3), F u T represents the resultant force in its longitudinal direction. r This refers to the steering torque caused by the speed difference between the two propellers in the propulsion structure.

[0044] 2. Further simplifying the overall hydrodynamic model and incorporating ocean current disturbances, the following kinematic and hydrodynamic mathematical models are obtained:

[0045]

[0046] In equation (5), τ u and τ r These represent the longitudinal thrust and the turning moment experienced by the unmanned vessel, respectively.

[0047] 3. Establish the mathematical description of the overall hydrodynamic model.

[0048]

[0049] In equation (6), x G τ represents the longitudinal component of the ship's center of gravity. X τ represents the longitudinal net external force. Y τ represents the resultant external force in the lateral direction. N This indicates the resultant deflection torque.

[0050] 4. By introducing a linear transformation formula, the overall hydrodynamic model is linearized to obtain the hydrodynamic model of the unmanned vessel.

[0051] By introducing the linearized transformation shown in equation (7), and using the mathematical description of the overall hydrodynamic model, the hydrodynamic model of the unmanned vessel is obtained as follows: equation (8):

[0052] u=u0m / s, v=0m / s, r=0deg / s,τ X =τ X0 N,τ N =0 N·m (7)

[0053]

[0054] In equations (7) and (8), I z Let u0 and τ represent the moments of inertia. X0 These are the initial values ​​set.

[0055] Based on the established hydrodynamic model of the unmanned vessel, the H-PINN method is introduced into the hydrodynamic model to obtain a simplified hydrodynamic model of the unmanned vessel. For details, see [link to details]. Figure 3 As shown, it includes the following steps:

[0056] Step S111: Embed the hydrodynamic model into the loss function of the H-PINN method to construct the loss function.

[0057]

[0058] The hydrodynamic coefficients in the above loss function expression are:

[0059]

[0060] In equation (9), L total L represents the total error value.u L v and L r These represent the error values ​​in the u, v, and r dimensions, respectively.

[0061] Step S112: Embed the Swish adaptive activation function into the H-PINN method.

[0062] The Swish adaptive activation function is embedded into the H-PINN method as the activation function for solving partial differential equations, and its expression is as follows:

[0063] f(x)=x·sigmoid(βx) (11)

[0064] In equation (11), the symbol β represents an adjustable parameter. If the condition β = 0 is met, the Swish function will be converted into a linear function, with the expression f(x) = 2 / x; if the condition β = ∞ is met, the Swish function will be converted into 0 or x. Therefore, the Swish function is a smooth function, belonging to an intermediate state between ReLU and linear functions. The first derivative of the Swish function conforms to the following relationship:

[0065] f′(x)=βf(x)+sigmoid(βx)(1-βf(x)) (12)

[0066] Step S113: Using the weight initialization method, the H-PINN method is used to solve the partial differential equation of the hydrodynamic coefficients in the constructed loss function.

[0067] The H-PINN method effectively solves the partial differential equations for hydrodynamic coefficients using a weight initialization method, namely the Xavier initialization method. The uniform distribution function is as follows:

[0068]

[0069] In equation (13), w represents the initial weights; U represents the distribution over the interval; n j The number of input neurons in the current layer; n j+1 The number of output neurons in the current layer.

[0070] Furthermore, the Adam optimization method is employed to optimize the gradient calculation of the loss function by dynamically adjusting the learning rate, and a minimum-maximum normalization function is constructed to quantitatively describe the accuracy of the hydrodynamic coefficient prediction results. The normalization function expression is as follows:

[0071]

[0072] Step S114: Based on the solution results of the hydrodynamic coefficients and the hydrodynamic model, construct a simplified hydrodynamic model of the unmanned vessel.

[0073] Step S12: Based on the simplified hydrodynamic model, a prediction model for unmanned vessel trajectory tracking is obtained. The model is designed using differential flatness theory to predict the desired trajectory and control quantity of the MPC. Stability analysis is designed based on the Lyapunov function.

[0074] See Figure 4 The diagram shown is a schematic of the model predictive control principle.

[0075] The prediction model for unmanned vessel trajectory tracking, derived from a simplified hydrodynamic model, can include the following steps 1-4:

[0076] 1. By discretizing and linearizing the simplified hydrodynamic model using the Model Prediction (MPC) method, the state-space model of the unmanned vessel is obtained as follows:

[0077]

[0078] Here, function f is a motion model equation representing the USV.

[0079] 2. Place the unmanned surface vessel model at any point (ζ) on the desired trajectory. d ,μ d The first-order Taylor expansion at ) is:

[0080]

[0081] In equation (16), ζ d ,μ d This indicates the expected trajectory (x) according to the USV. d ,y d The corresponding expected state variables and expected control variables are derived from this.

[0082] 3. Based on the state-space model and the first-order Taylor expansion, the error state-space differential equation of the linearized system is obtained:

[0083]

[0084] In equation (17), H and D are both Jacobian matrices.

[0085] By processing the above state-space equations using the forward Euler method, we obtain:

[0086]

[0087] In equation (18), H k =HT+I,D k =DT, where I is the identity matrix.

[0088] 4. The error state-space differential equation is transformed into the error state-space difference equation to obtain the prediction model for unmanned vessel trajectory tracking:

[0089] Ε(k)=Ξδ(k)+ΘΔU(19)

[0090] In equation (19), Ε(k) represents the prediction time domain N. P The predicted state variable sequence within the time domain, δ(k) represents the state error vector, and ΔU represents the control time domain N. C The sequence of control increments is defined within the given information, and Ξ and Θ represent the corresponding prediction system matrices. Their specific definitions are as follows:

[0091]

[0092] 5. Construct the objective function as follows:

[0093]

[0094] In equation (21), the first term on the right side represents the tracking capability, which refers to the desired trajectory. For the USV system, this means that the target trajectory can get closer and closer. The input cost of control is the second term on the right side. This equation also represents the USV, which hopes that its control quantity will remain stable in terms of change. The third term on the right side represents the terminal penalty function of the MPC controller. A, B, and G are the weight matrices of the controller.

[0095] 6. To ensure the closed-loop stability of the unmanned surface vessel system, the terminal penalty function is constructed as follows:

[0096]

[0097] In equation (22), θ>0 is the gain coefficient, A * =A+Γ T BΓ.

[0098] 7. The system constraints are designed as follows:

[0099]

[0100] In equation (23), N C It controls the time domain, ΔU min ΔU max U min U max Both are sets of minimum and maximum values, and are in N. C Within, the first two control the increment, while the latter two control the quantity, χ. min ,χ max The state variable in the prediction time domain N PThe set of minimum and maximum values ​​within a range is defined as follows:

[0101]

[0102] In equation (24), Indicates the Kronecker product. The number of rows is N C unit column vector, It is an identity matrix with the number of rows as the control dimension.

[0103] 8. Using the differential flatness theory, the model is designed to predict the desired trajectory and control quantity of the unmanned vessel controlling the MPC, as shown in equations (25) and (26), respectively:

[0104]

[0105] In equations (25) and (26), u d V represents the desired longitudinal velocity. d Represents the desired lateral velocity, r d Indicates the desired turning angular velocity. Indicates the desired angular velocity of the heading. Indicates the desired longitudinal acceleration. F1 represents the desired bow angle acceleration, and F2 represents the thrust of the left and right thrusters, respectively. and Let represent the first derivative of the thrust of the left and right thrusters with respect to time, respectively. and represents the second derivative of the thrust of the left and right thrusters with respect to time, respectively. The desired longitudinal acceleration and angular acceleration are combinations of the left and right thrusts, used to balance the thrusters and meet the motion requirements of the unmanned surface vessel on the desired trajectory. This refers to the control quantity at the desired longitudinal velocity; This is the control quantity at the desired turning angular velocity.

[0106] 9. Based on the stability analysis of the Lyapunov function, the objective function at time k+1 is:

[0107]

[0108] In equation (27),

[0109] Suppose that an optimal solution exists at time k+1. If the optimal solution can appear at time k+1, according to the MPC principle, this solution is no greater than the feasible solution, then:

[0110] J * (k+1)≤J(k+1)≤J * (k)(28)

[0111] That is, the objective function J * (k) is monotonically non-increasing, and at any time J has * (k)≥0.

[0112] The unmanned vessel trajectory tracking control method based on improved model predictive control provided in this invention utilizes differential flatness theory to design the desired trajectory and control quantity that conforms to the unmanned vessel system model, and applies these variables to model predictive control to improve trajectory tracking accuracy. In addition, to address the control quantity overshoot problem caused by the large deviation between the initial position and the desired trajectory of the unmanned vessel trajectory tracking, Lyapunov function optimization analysis is combined to guide the tracking of the desired trajectory, effectively solving the problem of large deviation between the initial and desired state quantities, reducing the amplitude of control quantity changes, and making speed changes smoother during navigation.

[0113] The unmanned surface vessel (USV) trajectory tracking control method based on improved model predictive control provided in this invention introduces the H-PINN method into the USV's hydrodynamic model and uses a physics-driven deep learning method instead of a data-driven deep learning method, resulting in a more accurate USV dynamic model. See also... Figure 5 The image shown is an example of the H-PINN method's prediction of the motion attitude of an unmanned surface vessel.

[0114] Based on the inventive concept of this invention, embodiments of this invention also provide an unmanned vessel trajectory tracking control device based on improved model predictive control, the structure of which is as follows: Figure 6 As shown, it includes:

[0115] Model building module 61 is used to introduce the H-PINN method into the hydrodynamic model of the unmanned vessel to obtain a simplified hydrodynamic model of the unmanned vessel.

[0116] The model predictive control optimization module 62 is used to obtain a predictive model for unmanned vessel trajectory tracking based on the simplified hydrodynamic model, design the desired trajectory and control quantity of model predictive control (MPC) using differential flatness theory, and design stability analysis based on Lyapunov function.

[0117] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0118] Based on the inventive concept of the present invention, embodiments of the present invention also provide a computer storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-mentioned unmanned vessel trajectory tracking control method based on improved model predictive control.

[0119] Based on the inventive concept of this invention, this embodiment of the invention also provides a server, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-mentioned unmanned vessel trajectory tracking control method based on improved model predictive control.

[0120] Unless otherwise specifically stated, terms such as processing, calculation, operation, determination, display, etc., may refer to the actions and / or processes of one or more processing or computing systems or similar devices that represent the manipulation and conversion of data representing physical (e.g., electronic) quantities within the registers or memory of the processing system into other data similarly representing physical quantities within the memory, registers, or other such information storage, transmission, or display devices of the processing system. Information and signals can be represented using any of a variety of different techniques and methods. For example, data, instructions, commands, information, signals, bits, symbols, and chips mentioned throughout the above description can be represented by voltage, current, electromagnetic waves, magnetic fields or particles, light fields or particles, or any combination thereof.

[0121] It should be understood that the specific order or hierarchy of steps in the disclosed process is an example of an exemplary method. Based on design preferences, it should be understood that the specific order or hierarchy of steps in the process may be rearranged without departing from the scope of this disclosure. The appended method claims provide elements of various steps in an exemplary order and are not intended to limit the scope to the specific order or hierarchy described.

[0122] In the detailed description above, various features are combined together in a single embodiment to simplify this disclosure. This approach to disclosure should not be construed as reflecting an intention that embodiments of the claimed subject matter require more features than those stated in each claim. Rather, as reflected in the appended claims, the invention is presented with fewer features than all of the features in a single disclosed embodiment. Therefore, the appended claims are hereby clearly incorporated into the detailed description, wherein each claim stands alone as a preferred embodiment of the invention.

[0123] Those skilled in the art will also understand that the various illustrative logic blocks, modules, circuits, and algorithm steps described in conjunction with the embodiments herein can be implemented as electronic hardware, computer software, or a combination thereof. To clearly illustrate the interchangeability between hardware and software, the various illustrative components, blocks, modules, circuits, and steps described above are generally described in terms of their functionality. Whether such functionality is implemented as hardware or software depends on the specific application and the design constraints imposed on the overall system. Those skilled in the art can implement the described functionality in alternative ways for each specific application; however, such implementation decisions should not be construed as departing from the scope of this disclosure.

[0124] The steps of the methods or algorithms described in conjunction with the embodiments herein can be directly embodied in hardware, software modules executed by a processor, or a combination thereof. The software modules can reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium well known in the art. An exemplary storage medium is connected to the processor, enabling the processor to read information from and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in a user terminal. Alternatively, the processor and storage medium can exist as discrete components in the user terminal.

[0125] For software implementation, the techniques described in this application can be implemented using modules (e.g., procedures, functions, etc.) that perform the functions described in this application. This software code can be stored in memory units and executed by a processor. The memory units can be implemented within the processor or outside the processor; in the latter case, they are communicatively coupled to the processor via various means, as is well known in the art.

[0126] The foregoing description includes examples of one or more embodiments. It is certainly impossible to describe all possible combinations of components or methods in order to describe the above embodiments, but those skilled in the art will recognize that further combinations and arrangements of the various embodiments are possible. Therefore, the embodiments described herein are intended to cover all such changes, modifications, and variations that fall within the scope of the appended claims. Furthermore, the term "comprising" as used in the specification or claims is interpreted in a manner similar to the term "including," as interpreted when used as a conjunction in the claims. Additionally, the use of any term "or" in the specification of the claims is intended to mean "non-exclusive or."

Claims

1. A trajectory tracking control method for unmanned surface vessels based on improved model predictive control, characterized in that, include: By introducing the H-PINN method into the hydrodynamic model of the unmanned vessel, a simplified hydrodynamic model of the unmanned vessel is obtained. Based on the simplified hydrodynamic model, a prediction model for unmanned vessel trajectory tracking is obtained. The model is designed using differential flatness theory to predict the desired trajectory and control variable of the MPC. Stability analysis is designed based on the Lyapunov function.

2. The method as described in claim 1, characterized in that, The hydrodynamic model is pre-constructed through the following steps: Construct an overall hydrodynamic model of the unmanned vessel; By introducing a linearization transformation, the overall hydrodynamic model is linearized to obtain the hydrodynamic model of the unmanned vessel.

3. The method as described in claim 2, characterized in that, The construction of the overall hydrodynamic model of the unmanned vessel includes: The overall hydrodynamic model of the unmanned vessel is constructed as follows (1): In equation (1), u represents the longitudinal velocity in the body coordinate system. Let v represent the longitudinal acceleration, and v represent the lateral velocity in the body coordinate system. 'r' represents lateral acceleration, and 'r' represents the turning angular velocity. τ represents the angular acceleration of the bow turn. u and τ r d represents the longitudinal thrust and turning moment experienced by the unmanned surface vessel, respectively; 11 =-X u -X |u|u |u|,d 22 =-Y v -Y |v|v |v|,d 33 =-N r -N |r|r |r|,, m (·) and d (·) These are all intermediate parameters, X (·) Y (·) and N (·) All belong to the hydrodynamic coefficients defined by SNAME, where m represents the mass of the hull, and I... ZZ This represents the moment of inertia of the bow roll. The mathematical description of the overall hydrodynamic model is as follows: (2) In equation (2), x G τ represents the longitudinal component of the ship's center of gravity. X τ represents the longitudinal net external force. Y τ represents the resultant external force in the lateral direction. N This indicates the resultant deflection torque.

4. The method as described in claim 3, characterized in that, The introduction of a linearization transformation to linearize the overall hydrodynamic model and construct an unmanned vessel hydrodynamic model includes: By introducing the linearized transformation shown in equation (3), and using the mathematical description of the overall hydrodynamic model, the hydrodynamic model of the unmanned vessel is obtained as follows: equation (4): u=u0,v=0,r=0,τ X =t X0 ,t N =0 (3) In equations (3) and (4), I z Let μ0 and τ represent the moments of inertia. X0 These are the initial values ​​set.

5. The method as described in claim 4, characterized in that, The process of introducing the H-PINN method into the hydrodynamic model of an unmanned vessel to obtain a simplified hydrodynamic model includes: The unmanned vessel hydrodynamic model is embedded into the loss function of the H-PINN method to construct the loss function; Embed the Swish adaptive activation function into the H-PINN method; By using a weight initialization method, the partial differential equations of the hydrodynamic coefficients in the constructed loss function can be solved using the H-PINN method. Based on the solution results of the hydrodynamic coefficients and the hydrodynamic model, a simplified hydrodynamic model of the unmanned vessel is constructed.

6. The method as described in claim 5, characterized in that, The construction of the loss function includes: The expression for constructing the loss function is as follows: The hydrodynamic coefficient in the loss function expression is: In equation (5), L total L represents the total error value. u L v and L r These represent the error values ​​in the u, v, and r dimensions, respectively.

7. The method as described in claim 1, characterized in that, The prediction model for unmanned vessel trajectory tracking based on the simplified hydrodynamic model includes: The simplified hydrodynamic model is discretized and linearized to obtain the state-space model of the unmanned vessel; The state-space model is expanded using a first-order Taylor series at any point on the desired trajectory. Based on the state-space model and the first-order Taylor expansion results, the error state-space differential equation is obtained. The error state space differential equation is transformed into an error state space difference equation to obtain a prediction model for unmanned vessel trajectory tracking.

8. The method as described in claim 7, characterized in that, The prediction model for obtaining the unmanned vessel trajectory tracking includes: The predictive model for unmanned surface vessel trajectory tracking is as follows: E(k)=Ξδ(k)+ΘΔU (7) In equation (7), E(k) represents the prediction time domain N. P The predicted state variable sequence within the time domain, δ(k) represents the state error vector, and ΔU represents the control time domain N. C The sequence of control increments within the range, where Ξ and Θ represent the corresponding prediction system matrices.

9. The method as described in claim 3, characterized in that, The method of using differential flatness theory to design a model to predict the desired trajectory and control variable of the control MPC includes: Using differential flatness theory, the model is designed to predict the desired trajectory and control variables of the unmanned vessel under MPC control, as shown in equations (8) and (9), respectively: In equations (8) and (9), u d V represents the desired longitudinal velocity. d v represents the desired lateral velocity. d Indicates the desired turning angular velocity. Indicates the desired heading angle. Indicates the desired longitudinal acceleration. F1 represents the desired bow angle acceleration, and F2 represents the thrust of the left and right thrusters, respectively. and Let represent the first derivative of the thrust of the left and right thrusters with respect to time, respectively. and Let represent the second derivatives of the thrust of the left and right thrusters with respect to time, respectively. This represents the control quantity at the desired longitudinal velocity; This represents the control quantity at the desired turning angular velocity.

10. A trajectory tracking control device for an unmanned surface vessel based on improved model predictive control, characterized in that, include: The model building module is used to introduce the H-PINN method into the hydrodynamic model of the unmanned vessel to obtain a simplified hydrodynamic model of the unmanned vessel. The model predictive control optimization module is used to obtain a predictive model for unmanned vessel trajectory tracking based on the simplified hydrodynamic model, design the desired trajectory and control quantity of the model predictive control (MPC) using differential flatness theory, and design stability analysis based on the Lyapunov function.

11. A computer storage medium, characterized in that, The computer storage medium stores computer-executable instructions, which, when executed by a processor, implement the unmanned vessel trajectory tracking control method based on improved model predictive control as described in any one of claims 1 to 9.

12. A server, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the unmanned vessel trajectory tracking control method based on improved model predictive control as described in any one of claims 1 to 9.

Images