Unmanned ship tracking control method for ecological breeding of shrimps and crabs in pond

Abstract

The invention discloses an unmanned ship tracking control method for ecological breeding of a shrimp and crab pond, and belongs to the technical field of aquaculture. The method comprises the following steps: firstly, establishing a three-degree-of-freedom kinetic model of an unmanned ship; by introducing hand coordinate transformation, an original under-actuated system is converted into a full-drive form; estimating unknown items in the dynamical model by using a time delay estimation technology, wherein the unknown items comprise deviation between an inertia matrix estimation value and an actual value, an unmodeled part in the dynamical model and external disturbance; an estimation error is suppressed and compensated by using a super-spiral fast terminal sliding mode control strategy, and a control law is designed; adaptive time delay estimation is further introduced to further improve the accuracy of disturbance estimation; the problems that an existing unmanned ship trajectory tracking control method depends on model parameters and is weak in disturbance rejection capacity are solved, accurate trajectory tracking control over the underactuated unmanned ship is achieved, the convergence speed is high, robustness is high, and a reference method is provided for trajectory tracking control over the underactuated unmanned ship.

Description

Technical Field

[0001] This invention relates to the field of unmanned vessel automatic navigation and control technology, specifically to an underactuated unmanned vessel trajectory tracking control method based on time delay estimation and sliding mode strategy for shrimp and crab pond ecological aquaculture, belonging to the field of aquaculture technology. Background Technology

[0002] In shrimp and crab pond aquaculture ecosystems, unmanned surface vessels (USVs) frequently need to traverse net cage arrays, avoid aeration equipment, and perform feeding and water quality sampling tasks along preset paths. The unique characteristics of the aquaculture environment present multiple challenges to control algorithms. High-precision trajectory tracking technology is crucial for achieving multi-task collaborative autonomous operation. However, the dynamics model of underactuated USVs exhibits nonlinearity and strong coupling characteristics, and is susceptible to model uncertainties, external disturbances such as water flow and waves, making it difficult for traditional control methods to balance accuracy and robustness. Existing methods such as model predictive control, backstepping control, and sliding mode control perform well under ideal conditions, but their common drawback lies in their dependence on precise dynamic models. Establishing such models requires a combination of theoretical derivation, numerical simulation, and experimental calibration, resulting in high computational complexity and economic costs. Furthermore, aquatic algae adhering to the hull after long-term operation significantly alter the surface friction coefficient and hydrodynamic characteristics, causing traditional algorithms based on fixed model parameters to frequently become unstable when the equipment load dynamically changes.

[0003] Time delay estimation techniques, by utilizing the time delay information from the previous sampling moment to estimate complex nonlinear terms in robot dynamics, effectively overcome the limitations of the aforementioned control methods. Compared to traditional methods, time delay estimation techniques have outstanding advantages such as simple algorithm structure, high real-time computational efficiency, and strong robustness, making them particularly suitable for complex systems with unmodeled dynamics and parameter uncertainties. However, the time delay estimation error is a state-dependent nonlinear term, the magnitude of which affects the system's robustness. Sliding mode control, with its unique variable structure characteristics and insensitivity to matched disturbances, can effectively suppress the adverse effects of time delay estimation residuals. Traditional time delay estimation techniques only consider the time delay information from the previous moment, resulting in poor estimation performance when dealing with intermittent disturbances, affecting system stability. Adaptive time delay estimation strategies are an improvement on traditional time delay estimation techniques, exhibiting good control performance under external time-varying disturbances. Existing research on time delay estimation techniques, both domestically and internationally, is based on devices such as robotic arms and underwater vehicles. There is still a research gap in the field of underactuated unmanned vessel control, especially in scenarios with complex terrain and time-varying disturbances in aquaculture waters. There is an urgent need to develop high-precision trajectory tracking control methods with low model dependence. Summary of the Invention

[0004] The purpose of this invention is to overcome the problems of existing unmanned surface vessel (USV) trajectory tracking control methods, which rely on model parameters and have weak disturbance resistance. It proposes a composite control method for USVs used in shrimp and crab pond aquaculture, based on adaptive time delay estimation and superspiral fast terminal sliding mode. This method requires no dynamic model parameters. It estimates unknown dynamic terms and external disturbances in real time through adaptive time delay estimation technology, and combines this with a superspiral fast terminal sliding mode strategy to compensate for estimation errors and suppress chattering, achieving high-precision and robust trajectory tracking control.

[0005] The objective of this invention is achieved through the following technical solutions.

[0006] A method for tracking and controlling unmanned vessels in shrimp and crab pond aquaculture includes the following steps:

[0007] Step 1: For the underactuated unmanned surface vessel, establish its underactuated mathematical model considering the three degrees of freedom of pitch, sway, and yaw:

[0008] (1);

[0009] Step 2: Introduce hand position coordinate transformation and reconstruct the kinematic model through state expansion.

[0010] Step 2.1, define the extended state vector: (2).

[0011] Step 2.2, for Taking the second derivative yields the fully actuated form of the dynamic model of the underactuated unmanned vessel:

[0012] (3);

[0013] Step 3: Use the time delay information from the previous sampling time to estimate the unknowns in the robot dynamics model, thus achieving controller design without the need for dynamics model parameters.

[0014] Step 3.1, based on the transformation of the hand position coordinates, the control input formula (3) is rewritten as:

[0015] (4);

[0016] In the formula: This is an estimate of the inertia matrix. This is the total disturbance.

[0017] Step 3.2, use time delay estimation techniques to... Estimating, we have:

[0018] (5);

[0019] Step 3.3, introduce the weight coefficient matrix ,reduce The effect on the controller is:

[0020] (6);

[0021] Step 4: Design a controller by using a superspiral fast terminal sliding mode control strategy to suppress and compensate for estimation errors.

[0022] Step 4.1, define the underactuated unmanned surface vessel tracking error as:

[0023] (7);

[0024] Step 4.2, select the second-order rapid terminal sliding surface:

[0025] (8);

[0026] Step 4.3: Differentiate equation (8) and substitute equations (6) and (7) to obtain the equivalent control term. :

[0027] (9);

[0028] Step 4.4, in order to reduce chattering and accelerate the system state to reach the terminal sliding surface, the super-spiral sliding mode switching control law is designed as follows:

[0029] (10);

[0030] Step 4.5, combine equations (9) and (10) to obtain the overall control law:

[0031] (11);

[0032] Step 5: By introducing adaptive time delay estimation, the dynamic estimate is adaptively calculated based on the measurement gradient of the sliding mode variable, further improving the accuracy of the time delay estimation in Step 3.

[0033] Step 5.1 introduces an adaptive model estimation strategy based on sliding mode variables, resulting in an improved bias estimate. The expression is:

[0034] (12);

[0035] Where: Adaptive weight matrix The adjustment is adaptive based on the gradient of the sliding mode variable, and the specific expression is as follows:

[0036] (13);

[0037] In the formula: (•) i and(••) ii These are the first and second digits of the vector (•). The nth element and the diagonal matrix (••) One diagonal element; and It is a positive adaptive gain, used to change The rate; It is a non-negative constant, mainly used to ensure the stability of the adaptive estimation strategy.

[0038] From equation (13), we can see that It will change with the sliding mode variables, and when the variables converge to the sliding surface, Only then will it converge, thereby achieving [the desired outcome]. More effective dynamic estimation.

[0039] Step 5.2, using equation (12) In the substitution formula (11) The overall control law is obtained as follows:

[0040] (14); Attached Figure Description

[0041] Figure 1 This is a flowchart of the algorithm of the present invention.

[0042] Figure 2 This is a schematic diagram of the planar motion coordinate system of an underactuated unmanned surface vessel.

[0043] Figure 3 This is a structural block diagram of the controller of the present invention. Detailed Implementation

[0044] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. The described examples are only some embodiments of the present invention, not all embodiments.

[0045] The specific implementation steps are as follows:

[0046] Step 1: Establish an underactuated unmanned surface vessel (USV) dynamics model for the multi-rotor aerodynamic USV, such as... Figure 2 As shown in the figure Indicates an inertial coordinate system. Representing the body coordinate system, and based on the inertial coordinate system and the body coordinate system, considering the three degrees of freedom of motion (sway, roll, and bow), the following mathematical model is established:

[0047] (1);

[0048] Step 2: Introduce hand position point coordinate transformation, reconstruct the kinematic model through state extension, and obtain equivalent control input for heading.

[0049] Step 2.1, define the extended state vector:

[0050] (2);

[0051] Step 2.2, for Taking the second derivative yields the fully actuated form of the dynamic model of the underactuated unmanned vessel:

[0052] (3);

[0053] Step 3, as follows Figure 3 As shown, time delay estimation technology is used to estimate the unknown terms and external disturbance terms in the dynamic model, so as to realize the controller design without dynamic model parameters.

[0054] Step 3.1, based on the transformation of the hand position coordinates, the control input formula (3) is rewritten as: (4);

[0055] Step 3.2, use time delay estimation techniques to... Estimating, we have:

[0056] (5);

[0057] Step 3.3, introduce the weight coefficient matrix ,reduce The effect on the controller is:

[0058] (6);

[0059] Step 4, as follows Figure 3 As shown, the superspiral fast terminal sliding mode control strategy is used to suppress and compensate for estimation errors. Fast terminal sliding mode control is used to reduce steady-state errors and improve dynamic performance. The superspiral sliding mode algorithm is used to accelerate the convergence of the system state to the sliding surface and reduce chattering.

[0060] Step 4.1, define the underactuated unmanned surface vessel tracking error as:

[0061] (7);

[0062] Step 4.2, select the second-order rapid terminal sliding surface:

[0063] (8);

[0064] Step 4.3: Differentiate equation (8) and substitute equations (6) and (7) to obtain the equivalent control term. :

[0065] (9);

[0066] Step 4.4, in order to reduce chattering and accelerate the system state to reach the terminal sliding surface, the super-spiral sliding mode switching control law is designed as follows:

[0067] (10);

[0068] Step 4.6: Combine equations (9) and (10) to obtain the overall control law:

[0069] (11);

[0070] Step 5, as follows Figure 3 As shown, based on the measurement gradient of the sliding mode variable, the total disturbance is adaptively estimated, which further improves the accuracy of the time delay estimation in step 3, and thereby improves the controller.

[0071] Step 5.1, Improved deviation estimate The expression is:

[0072] (12);

[0073] Where: Adaptive weight matrix The adjustment is adaptive based on the gradient of the sliding mode variable, and the specific expression is as follows:

[0074] (13);

[0075] In the formula: (•) i and(••) ii These are the first and second digits of the vector (•). The nth element and the diagonal matrix (••) One diagonal element; and It is a positive adaptive gain, used to change The rate; It is a non-negative constant, mainly used to ensure the stability of the adaptive estimation strategy.

[0076] From equation (13), we can see that It will change with the sliding mode variables, and when the variables converge to the sliding surface, Only then will it converge, thereby achieving [the desired outcome]. More effective dynamic estimation.

[0077] Step 5.2, using equation (12) In the substitution formula (11) The overall control law is obtained as follows:

[0078] ;

[0079] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the technical scope of the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. An unmanned ship tracking control method for shrimp and crab pond ecological breeding, the method mainly comprises the following 5 steps: Step 1: For the problem of accurate trajectory tracking of underactuated unmanned ships in shrimp and crab pond ecological breeding scenes, a three-degree-of-freedom mathematical model is established, Step 2: Hand coordinate transformation is introduced to obtain equivalent control input for the bow, and the original underactuated system is converted into a fully driven form; Step 3: Use time delay estimation technology to estimate unknown terms and external disturbance terms in the dynamic model; Step 4: Use super-spiral fast terminal sliding mode control strategy to suppress and compensate estimation error; Step 5: And by introducing adaptive time delay estimation, based on the measurement gradient of the sliding variable, the dynamic estimation value is adaptively calculated, further improving the accuracy of time delay estimation in step 3.

2. The unmanned ship tracking control method for shrimp and crab pond ecological breeding according to claim 1, characterized in that: The step 1 specifically comprises: An underactuated unmanned ship dynamic model is established, based on the inertial coordinate system and the body coordinate system, considering three degrees of freedom motion of surge, sway and yaw, the following mathematical model is established: （1）； wherein: , and denote the position coordinates and the heading angle of the USV relative to the inertial coordinate system, respectively; , , denote the longitudinal, lateral velocity and the yaw angular velocity in the body coordinate system, respectively; , denote the longitudinal thrust and the yaw moment acting on the USV, respectively; , , denotes the inertial mass of the USV; denotes the longitudinal hydrodynamic parameter related to the longitudinal acceleration, denotes the lateral hydrodynamic parameter related to the lateral acceleration, denotes the yaw hydrodynamic parameter related to the yaw angular acceleration, denotes the mass of the USV; denotes the moment of inertia; denotes the longitudinal hydrodynamic parameter related to the longitudinal velocity, denotes the lateral hydrodynamic parameter related to the lateral velocity, denotes the yaw hydrodynamic parameter related to the yaw angular velocity; denotes the longitudinal hydrodynamic parameter related to the square of the longitudinal velocity, denotes the lateral hydrodynamic parameter related to the square of the lateral velocity, denotes the yaw hydrodynamic parameter related to the square of the yaw angular velocity; , , denotes the unknown terms such as the unmodeled parts, the parameter perturbation and the external disturbance, etc.

3. The unmanned ship tracking control method for shrimp and crab pond ecological breeding according to claim 1, characterized in that, The step 2 specifically comprises: Hand position point coordinate transformation is introduced, and the kinematic model is reconstructed by state extension to make the bow obtain equivalent control input; The extended state vector is defined as: （2）； wherein: , is the coordinate of the USV center of mass in the inertial coordinate system, , is the horizontal and vertical coordinate of the hand position point; is the length of the hand position point from the center of mass of the USV. For Taking the second derivative of the dynamics model of the underactuated USV gives the fully actuated form: （3）； In the formula: and is a state variable, is a control input, , , , The specific expression of is: and and and （4）； The above expression can be specifically expressed as: and and (4); with is a nonsingular matrix, the unknowns are denoted by the components of the vector and let the unknowns be denoted by the components of the vector with are bounded.

4. The unmanned ship tracking control method for shrimp and crab pond ecological breeding according to claim 1, characterized in that, The step 3 specifically comprises: Use the time delay information at the previous sampling time to estimate the unknown terms in the robot dynamic model, realize the controller design without the need of dynamic model parameters; Based on the hand position point coordinate transformation, the control input (3) is rewritten as (5); wherein is the estimated value of the inertia matrix, i.e. the total disturbance, ; Using the time delay estimation technique to estimate Then we have: (6); wherein: (•) t denotes the current time state quantity (•) t -L denotes the previous sample time state quantity (•) is the sample time interval, which is related to the sensor sampling frequency, the smaller the estimation value is the more accurate; acceleration of the system at a previous time may be obtained by second-order differentiation of the position signal: （6）； Introducing a weight coefficient matrix , reduces the influence on the controller, resulting in (7).

5. The unmanned ship tracking control method for shrimp and crab pond ecological breeding according to claim 1, characterized in that, The step 4 specifically comprises: Use super-spiral sliding mode algorithm (STA) to suppress and compensate estimation error, use fast terminal sliding mode control (FTSMC) to reduce steady-state error and improve dynamic performance, and use super-spiral sliding mode algorithm to speed up the system state to converge to the sliding mode surface and reduce chattering phenomenon; The tracking error of the underactuated USV is defined as: (8); A second-order fast terminal sliding surface is selected: (9); wherein: , and is a positive constant diagonal matrix, and is a positive constant to be set when the system deviates from the stable state, the term can accelerate the convergence speed of the system, and i.e. and there is no negative exponential term, avoiding the problem of non-singularity; Differentiate equation (9) and substitute equations (7) and (8) to obtain the equivalent control term : （10）； In order to weaken the chattering phenomenon and speed up the system state to reach the terminal sliding mode surface, the super-spiral sliding mode switching control law is designed as: （11）； wherein is a linear term added in the control law, aiming to accelerate the convergence speed, , , is a positive constant diagonal matrix The total control law is obtained by combining equation (10) and (11): （12）。 6. The unmanned ship tracking control method for shrimp and crab pond ecological breeding according to claim 1, characterized in that, The step 5 specifically comprises: Adaptive time delay estimation (ATDE) based on the measurement gradient of the sliding variable is proposed to adaptively estimate the total disturbance, further improve the accuracy of time delay estimation in step 3, and improve the controller. To further improve the accuracy of the total bias estimate The improved bias estimate is given by （13）； In the formula: adaptive weight matrix According to the gradient adaptive adjustment of the sliding mode variable, the specific expression is: （14）； where (•) and (••) are the first and second elements of the vector (•), respectively, and (•••) is the first diagonal element of the diagonal matrix (•••). i ii is a non-negative constant, mainly used to guarantee the stability of the adaptive estimation strategy.​​​​​​​ From equation (14), we can see that It will change with the sliding mode variables, and when the variables converge to the sliding surface, Only then will it converge, thereby achieving [the desired outcome]. More effective dynamic estimation; Instead of (12) in the formula (13) is used The total control law is obtained as​ （15）。

Images