Unmanned ship prediction tracking control method, device and equipment based on disturbance compensation and medium

Abstract

The invention provides an unmanned ship prediction tracking control method and device based on disturbance compensation, equipment and a medium, and the method comprises the steps: determining an unmanned ship model and a thrust value of a propeller module of an unmanned ship, determining a disturbance value of the propeller module according to the unmanned ship model and the thrust value of the propeller module, and carrying out the prediction tracking control of the unmanned ship. According to the radial basis function neural network and the thrust value and the disturbance value of the thruster module, a disturbance estimation value of the thruster module is obtained through prediction, and based on the disturbance estimation value, an unmanned ship model is simplified into a precisely known linear model, so that the thrust value of the thruster module is obtained according to the linear model and the thrust value of the thruster module. The target control quantity of the thruster module is predicted more accurately, the disturbance compensation output is determined according to the disturbance estimation value, the target control output of the thruster module is determined according to the target control quantity and the disturbance compensation output, the accuracy and stability of the target control output are guaranteed, and meanwhile complex manual parameter adjustment is avoided.

Description

Technical Field

[0001] The present application relates to the field of ocean monitoring, and in particular to a method, device, equipment and medium for predictive tracking control of an unmanned ship based on disturbance compensation. Background Art

[0002] Unmanned surface vessels currently have broad application prospects in areas such as marine environmental monitoring and water patrol. They are typically equipped with one or more thrusters to generate propulsion for driving. For unmanned vessels, their motion control involves two issues: path tracking and trajectory tracking. Path tracking is to control the unmanned vessel to follow a time-independent geometric path in space. It does not have strict time requirements and only requires that the motion position of the unmanned vessel coincide with the path position. The goal of trajectory tracking is to track a time-dependent reference trajectory. It has strict time requirements, that is, it needs to reach a specified position on the trajectory at a specified time, and it must meet both spatial and temporal constraints.

[0003] At present, for path tracking, most people use the line-of-sight method as the speed planner. The methods used in speed control are different. The most common ones are PID control, active disturbance rejection control, and model predictive control. The parameters of PID control and active disturbance rejection control usually need to be adjusted based on experience, and their parameters need to be adjusted repeatedly for different ships. Model predictive control requires the establishment of an accurate mathematical model. However, due to the complexity of unmanned ships and the uncertainty of external interference such as water flow, only an approximate model of the unmanned ship can be established as the input of the prediction model to calculate the propulsion device, resulting in large errors. Summary of the Invention

[0004] The present application provides a method, apparatus, device, and medium for predictive tracking control of an unmanned vessel based on disturbance compensation to address at least one problem existing in the related art. The technical solution is as follows:

[0005] In a first aspect, an embodiment of the present application provides a method for predictive tracking control of an unmanned vessel based on disturbance compensation, comprising:

[0006] Determining an unmanned ship model and a thrust value of a propulsion module of the unmanned ship, and determining a disturbance value of the propulsion module based on the unmanned ship model and the thrust value of the propulsion module;

[0007] According to the radial basis function neural network, the thrust value of the propeller module and the disturbance value, the disturbance estimation value of the propeller module is predicted;

[0008] Based on the disturbance estimation value, the unmanned ship model is simplified into a linear model. According to the linear model and the thrust value of the propeller module, the target control variable of the propeller module is predicted.

[0009] The disturbance compensation output is determined based on the disturbance estimation value, and the target control output of the thruster module is determined based on the target control amount and the disturbance compensation output.

[0010] In one embodiment, determining the unmanned ship model includes:

[0011] A kinematic model is constructed based on the position of the unmanned ship in the plane coordinate system, the derivative velocity of the position, the rotation velocity, the forward velocity of the unmanned ship along one coordinate axis and the lateral velocity along another coordinate axis in the unmanned ship coordinate system, and the angle between the plane coordinate system and the unmanned ship coordinate system;

[0012] Constructing a dynamic model based on the spacing between the first propeller and the second propeller in the propulsion module, the mass of the unmanned ship, the differential of the forward speed of the unmanned ship along one of the coordinate axes in the unmanned ship coordinate system, the differential of the rotational speed, the thrust of the first propeller, and the thrust of the second propeller;

[0013] Among them, the unmanned ship model includes a kinematic model and a dynamic model.

[0014] In one embodiment, the thrust value of the propeller module includes a current thrust value of the first propeller and a current thrust value of the second propeller; and calculating the disturbance value of the propeller module according to the unmanned ship model and the thrust value of the propeller module includes:

[0015] Obtain the actual values ​​of the current derivative speed and angle, and combine them with the kinematic model to calculate the current forward speed and the differential of the current forward speed;

[0016] The actual values ​​of the rotation speed and the mass of the unmanned ship are obtained, and combined with the dynamic model to determine the disturbance value of the propulsion module. The disturbance value of the propulsion module includes the resistance encountered by the unmanned ship in the direction of the forward speed during movement and the torque of the resistance acting on the center of gravity of the unmanned ship.

[0017] In one embodiment, predicting a disturbance estimate value of the thruster module based on the radial basis function neural network, the thrust value of the thruster module, and the disturbance value includes:

[0018] Constructing an input vector of the input layer of the radial basis function neural network based on the current thrust value of the first propeller, the current thrust value of the second propeller, the angle, the actual value of the rotation speed, and the current forward speed, and using the disturbance value of the propeller module as the output vector;

[0019] Determine an output vector of a hidden layer of the radial basis function neural network according to the radial basis function neural network, the input vector, and the output vector;

[0020] According to the output vector of the hidden layer and the weight matrix from the hidden layer to the output layer, the disturbance estimation value of the thruster module is predicted;

[0021] The disturbance estimation value includes a first estimation value of the resistance experienced by the unmanned ship in the direction of the forward speed during movement and a second estimation value of the moment magnitude of the first estimation value acting on the center of gravity of the unmanned ship.

[0022] In one embodiment, predicting the target control amount of the propulsion module based on the linear model and the thrust value of the propulsion module includes:

[0023] Based on the sampling time, the linear model is discretized to determine the discrete state equation;

[0024] Determine a prediction variable based on the forward speed, angle, and prediction step size, and determine a decision variable based on the control step size. The decision variable is a set of control variables of the thruster module at the current moment and several moments in the future. Each control variable includes the thrust of the first thruster and the thrust of the second thruster.

[0025] Determine the prediction model based on the discrete state equation, prediction variables and decision variables;

[0026] Determine the expected forward speed and expected angle of the unmanned vessel, and determine the expected output based on the expected forward speed, expected angle, and predicted step size;

[0027] Based on the performance function, expected output and prediction model, the optimal solution of the decision variable is determined, and the control amount of the thruster module corresponding to the current moment is selected as the target control amount from the optimal solution of the decision variable.

[0028] In one embodiment, determining the disturbance compensation output according to the disturbance estimate includes:

[0029] Obtaining a moment of inertia of the unmanned vessel, calculating a ratio of the moment of inertia to the distance, determining a first product of half the mass of the unmanned vessel and a first estimated value, and a second product of the ratio and a second estimated value;

[0030] Determining a first compensation output according to the inverse of a difference between the first product and the second product, and determining a second compensation output according to the inverse of a sum of the first product and the second product;

[0031] A disturbance compensation output is obtained according to the first compensation output and the second compensation output.

[0032] In one embodiment, determining the target control output of the thruster module according to the target control variable and the disturbance compensation output includes:

[0033] The target control quantity and the sum of the disturbance compensation output are calculated to obtain the target control output of the thruster module.

[0034] In a second aspect, an embodiment of the present application provides an unmanned ship prediction tracking control device based on disturbance compensation, comprising:

[0035] a first determining module, configured to determine an unmanned ship model and a thrust value of a propeller module of the unmanned ship, and determine a disturbance value of the propeller module according to the unmanned ship model and the thrust value of the propeller module;

[0036] A first prediction module is configured to predict a disturbance estimation value of the propulsion module based on a radial basis function neural network, a thrust value of the propulsion module, and a disturbance value;

[0037] A second prediction module is used to simplify the unmanned ship model into a linear model based on the disturbance estimation value, and predict the target control amount of the propulsion module according to the linear model and the thrust value of the propulsion module;

[0038] The second determination module is used to determine the disturbance compensation output according to the disturbance estimation value, and to determine the target control output of the thruster module according to the target control amount and the disturbance compensation output.

[0039] In a third aspect, an embodiment of the present application provides an electronic device, comprising: a processor and a memory, wherein the memory stores instructions, and the instructions are loaded and executed by the processor to implement the method in any one of the above-mentioned embodiments.

[0040] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores a computer program. When the computer program is executed, it implements the method in any one of the above-mentioned embodiments.

[0041] The beneficial effects of the above technical solution include at least:

[0042] By determining the thrust value of the unmanned ship model and the propeller module of the unmanned ship, and determining the disturbance value of the propeller module based on the unmanned ship model and the thrust value of the propeller module, the disturbance estimation value of the propeller module is predicted based on the radial basis function neural network, the thrust value of the propeller module and the disturbance value, and simplifying the unmanned ship model into an accurately known linear model based on the disturbance estimation value, the target control quantity of the propeller module is more accurately predicted based on the linear model and the thrust value of the propeller module, and the disturbance compensation output is determined based on the disturbance estimation value, and the target control output of the propeller module is determined based on the target control quantity and the disturbance compensation output, thereby ensuring the accuracy and stability of the target control output and avoiding manual complex parameter adjustment.

[0043] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, the above various aspects, various embodiments, and features will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the multiple drawings represent the same or similar components or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments disclosed in this application and should not be construed as limiting the scope of this application.

[0045] Figure 1 This is a flowchart of the steps of a predictive tracking control method for an unmanned ship based on disturbance compensation according to an embodiment of the present application;

[0046] Figure 2 This is a schematic diagram of the coordinate system involved in an unmanned boat according to an embodiment of the present application;

[0047] Figure 3 This is a schematic diagram of the path tracking of an unmanned boat according to an embodiment of the present application;

[0048] Figure 4 This is a structural block diagram of an unmanned ship prediction and tracking control device based on disturbance compensation according to an embodiment of the present application;

[0049] Figure 5 This is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0050] Hereinafter, only certain exemplary embodiments are briefly described. As will be appreciated by those skilled in the art, the described embodiments may be modified in various ways without departing from the spirit or scope of the present application. Therefore, the drawings and description are to be regarded as illustrative in nature and not restrictive.

[0051] Reference Figure 1 , a flowchart of a predictive tracking control method for an unmanned ship based on disturbance compensation according to an embodiment of the present application is shown. The predictive tracking control method for an unmanned ship based on disturbance compensation may include at least steps S100-S400:

[0052] S100: Determine an unmanned ship model and a thrust value of a propeller module of the unmanned ship, and determine a disturbance value of the propeller module according to the unmanned ship model and the thrust value of the propeller module.

[0053] S200: Predicting a disturbance estimation value of the thruster module based on the radial basis function neural network, the thrust value of the thruster module, and the disturbance value.

[0054] S300: Based on the disturbance estimation value, simplify the unmanned ship model into a linear model, and predict the target control amount of the propulsion module according to the linear model and the thrust value of the propulsion module.

[0055] S400 , determining a disturbance compensation output according to the disturbance estimation value, and determining a target control output of the thruster module according to the target control amount and the disturbance compensation output.

[0056] The technical solution of the embodiment of the present application determines the thrust value of the unmanned ship model and the propeller module of the unmanned ship, and determines the disturbance value of the propeller module based on the unmanned ship model and the thrust value of the propeller module. The disturbance estimation value of the propeller module is predicted based on the radial basis function neural network, the thrust value and the disturbance value of the propeller module. Based on the disturbance estimation value, the unmanned ship model is simplified into a precisely known linear model. Therefore, according to the linear model and the thrust value of the propeller module, the target control quantity of the propeller module is more accurately predicted. According to the disturbance estimation value, the disturbance compensation output is determined. According to the target control quantity and the disturbance compensation output, the target control output of the propeller module is determined, thereby ensuring the accuracy and stability of the target control output and avoiding manual complex parameter adjustment.

[0057] In one embodiment, the unmanned ship model includes a kinematic model and a dynamic model. Determining the unmanned ship model in step S100 includes steps S110-S120:

[0058] S110. Construct a kinematic model based on the position of the unmanned ship in the plane coordinate system, the derivative velocity of the position, the rotational velocity, the forward velocity of the unmanned ship along one of the coordinate axes in the unmanned ship coordinate system, the lateral velocity along the other coordinate axis, and the angle between the plane coordinate system and the unmanned ship coordinate system.

[0059] Alternatively, as Figure 2 As shown, the kinematic model of the unmanned ship in the embodiment of the present application takes a fixed double-propeller model as an example. The unmanned ship has a combined navigation module 1 and a propeller module. The propeller module includes a first propeller (also called a left propeller) 21 and a second propeller (also called a right propeller) 22. The distance between the first propeller 21 and the second propeller 22 is denoted as d, which can be measured in advance. When constructing the kinematic model, the origin is O. W Construct plane coordinate system W={X W ,Y W} and the unmanned ship coordinate system B={X B ,Y B}, the angle between the plane coordinate system W and the unmanned ship coordinate system B (also known as the heading angle of the unmanned ship), the position of the unmanned ship in the plane coordinate system W is p = {x, y}, where x is the horizontal coordinate and y is the vertical coordinate, and the derivative velocity of the position is The rotation speed r (i.e., the rotation speed r of the unmanned ship in the plane), the unmanned ship's position along one of the coordinate axes (Y B ) and the forward speed u along another coordinate axis (X B )'s lateral velocity v and angle Construct a kinematic model.

[0060] Specifically, take the state vector (pose) of the unmanned ship The speed of the unmanned ship in coordinate system B is expressed as The kinematic model of the unmanned ship can be obtained as:

[0061]

[0062] Right now:

[0063]

[0064] in, It can be understood that the unmanned ship moves along the X W The rate of change of position in the direction, It can be understood that the unmanned ship moves along the Y axis in the plane coordinate system W. W Directional position change rate, position p = {x, y}, derivative velocity It can be obtained based on the combined navigation module 1; Right now The differential of

[0065] S120. Construct a dynamic model based on the distance between the first propeller and the second propeller in the propeller module, the mass of the unmanned ship, the differential of the forward speed of the unmanned ship along one of the coordinate axes in the unmanned ship coordinate system, the differential of the rotational speed, the thrust of the first propeller, and the thrust of the second propeller.

[0066] Optionally, based on the distance d between the first propeller and the second propeller in the propeller module, the mass m of the unmanned ship, the position of the unmanned ship in the unmanned ship coordinate system B along one of the coordinate axes (Y B )'s forward velocity u Differential of rotational speed r The thrust of the first thruster f l and the thrust of the second thruster f r , construct the dynamic model:

[0067]

[0068] in, is the differential of the lateral velocity v, I is the moment of inertia of the unmanned ship (which can be obtained in advance), and the thrust value of the propeller module of the unmanned ship It can also be directly obtained based on the unmanned ship system, and the current thrust value of the first propeller is obtained and substituted into f l , the current thrust value of the second thruster is substituted into f r ;w u is the resistance that the unmanned ship encounters in the direction of forward speed u during movement, w r is the torque of the resistance acting on the center of gravity of the unmanned ship, w v is the resistance experienced by the unmanned vessel in the direction of the lateral velocity v during its motion. It should be noted that the magnitude of these resistances is unknown and is equivalent to the resistance to the propulsion module. They may be caused by the structure of the hull itself and environmental factors such as water currents and waves. They are also considered disturbance values ​​and are estimated using a radial basis function neural network.

[0069] Among them, since the fixed double-propeller unmanned ship is an under-actuated unmanned ship, B There is no propulsion force in the direction, so w is not considered in subsequent calculations. v This unknown disturbance, that is, not considered Therefore, the disturbance value of the thruster module at this time includes w u and w r , the final dynamic model is written as:

[0070]

[0071] In one embodiment, determining the unmanned ship model in step S100 includes steps S130-S140:

[0072] S130 , obtaining actual values ​​of the current derivative speed and angle, and calculating the current forward speed and the differential of the current forward speed in combination with the kinematic model.

[0073] Optionally, get the current derivative velocity The actual value and angle After the actual value of , combined with the above kinematic model, the current forward speed u and the differential of the forward speed u are calculated. Specific values:

[0074]

[0075] S140: Obtain the actual values ​​of the rotation speed and the mass of the unmanned ship, and determine the disturbance value of the thruster module in combination with the dynamic model.

[0076] Similarly, the actual value of the rotation speed r and the actual value of the mass m of the unmanned ship are obtained, and the differentials of the rotation speed r and the forward speed u are calculated. Combined with the dynamic model, that is, formula (4), the disturbance value of the specific thruster module is calculated:

[0077]

[0078] In one embodiment, step S200 includes steps S210-S230:

[0079] S210. According to the current thrust value of the first propeller, the current thrust value of the second propeller, the angle, the actual value of the rotation speed and the current forward speed, an input vector of the input layer of the radial basis function neural network is constructed, and the disturbance value of the propeller module is used as the output vector.

[0080] It should be noted that the present application is based on the radial basis function neural network (RBF) disturbance observer for subsequent processing. The radial basis function neural network is a feedforward neural network based on the radial basis function, which is widely used in function approximation and nonlinear regression. Its network structure includes an input layer, a hidden layer, and an output layer. Optionally, according to the current thrust value of the first propeller (substituted into f l ), the current thrust value of the second thruster (substituted into f r ), angle The actual value of the rotation speed r and the current forward speed u are used to construct the input vector s of the input layer of the radial basis function neural network, and the disturbance value of the thruster module is used as the output vector

[0081]

[0082] S220 , determining an output vector of a hidden layer of the radial basis function neural network according to the radial basis function neural network, the input vector, and the output vector.

[0083] Optionally, the hidden layer activation function of the radial basis function neural network selects a Gaussian distribution function, and the number of nodes N in the hidden layer is designed based on autonomous regulation, where c j is the jth node center, φ j (s) is the calculation result corresponding to the center of the jth node, σ j is the width parameter of the distribution, and e is the natural logarithm:

[0084]

[0085] Thus, the output vector φ(x) of the hidden layer can be obtained:

[0086]

[0087] S230. Predict a disturbance estimate of the thruster module based on the output vector of the hidden layer and the weight matrix from the hidden layer to the output layer.

[0088] Optionally, when designing the output layer of the radial basis function neural network, for the disturbance observer, the purpose is to estimate the disturbance of the unmanned ship, so the output layer corresponds to the model with two nodes, namely the disturbance estimation value of the disturbance for:

[0089]

[0090] in, For w u The estimated value of For w r estimated value.

[0091] In the embodiment of the present application, the weight matrix W′∈R from the hidden layer to the output layer is taken as 2×N The dimension of the weight matrix W′ is related to the number of nodes in the hidden layer and the number of output nodes in the output layer. The value of the weight matrix W′ is calculated in real time through online updating after a given initial value. 1N Refers to the weight value at the 1st row and Nth column position.

[0092]

[0093] Therefore, based on the product of the output vector φ(s) of the hidden layer and the weight matrix W′ from the hidden layer to the output layer, the perturbation estimate of the output layer can be obtained:

[0094]

[0095] Among them, the disturbance estimate Includes a first estimate of the resistance experienced by the unmanned ship in the direction of forward speed during movement and a second estimated value of the moment magnitude of the first estimated value acting on the center of gravity of the unmanned ship

[0096] It should be noted that the update rule of the weight matrix W′ is defined as follows, where T is the transpose and η is the learning rate:

[0097]

[0098] Among them, W′ 更新前 Initially, the initial value of W' is given, and updates are started from the initial value.

[0099] In one embodiment, in step S300, the unmanned ship model is simplified into a linear model based on the disturbance estimation value, specifically:

[0100] Based on formula (10) and formula (4), in order to ensure that the disturbance during the motion of the unmanned ship does not interfere with the acceleration of the unmanned ship, it is necessary to calculate the disturbance estimation value required by the first propeller and the second propeller: Therefore, assuming that the compensation output of the compensator of the second thruster (ie, the second compensation output) is f rb , the compensation output of the first thruster (ie the first compensation output) is f lb , let the left side of formula (4) The following equation can be obtained:

[0101]

[0102] From this, the disturbance estimation values ​​required by the first thruster and the second thruster can be calculated as Compensated output in:

[0103]

[0104] At this point, the unmanned ship model can be simplified to a known linear model:

[0105]

[0106] In one embodiment, predicting the target control amount of the propulsion module based on the linear model and the thrust value of the propulsion module in step S300 includes steps S310-S350:

[0107] S310 : Discretize the linear model based on the sampling time and determine a discrete state equation.

[0108] Alternatively, assume that the sampling time is T s , then based on the linear model, the state variable function value at the next moment k+1 and the current moment k is as follows:

[0109]

[0110] Take the state vector at time k Output at time k The control quantity of the thruster module is At this time, the discretized discrete state equation can be written as:

[0111]

[0112] in, is the angle at time k, is the angle at time k+1, r(k) is the rotation speed at time k, r(k+1) is the rotation speed at time k+1, u(k) is the forward speed at time k, u(k+1) is the forward speed at time k+1, f r (k) is the thrust of the second thruster at time k, f l (k) is the thrust of the first thruster at time k, z(k+1) is the state vector at time k+1, g(k+1) is the output at time k+1, A(k), B(k), and C(k) are all results expressed in matrix form based on formula (15), where:

[0113]

[0114] S320 : Determine a prediction variable based on the forward speed, the angle, and the predicted step length, and determine a decision variable based on the control step length.

[0115] Optionally, based on the forward speed, angle and predicted step length N p Determine the prediction variable G(k+1), that is, the set of outputs g(k) at different times, and the control step size N c Determine the decision variable F(k), which is a set of control variables of the thruster modules at the current moment and several moments in the future, where each control variable includes the thrust of the first thruster and the thrust of the second thruster.

[0116]

[0117] S330: Determine a prediction model based on the discrete state equation, the prediction variables, and the decision variables.

[0118] Optionally, the calculation formula for the output of each step within the prediction step, that is, the prediction model is:

[0119] G(k+1)=C p (k)(A p (k)g(k)+B p (k)F(k)) (17)

[0120] in:

[0121]

[0122] Where i is an integer, That is, an all-zero matrix of dimension 2x3.

[0123] S340: Determine the expected forward speed and expected angle of the unmanned ship, and determine the expected output based on the expected forward speed, the expected angle, and the predicted step size.

[0124] Alternatively, as Figure 3As shown in the figure, the path tracking of the unmanned ship is shown. Assume that the path currently tracked by the unmanned ship is P k P k+1 , its horizontal coordinate is the coordinate expressed by the plane coordinate system W. The lateral error of the distance path of the unmanned ship is x d , take P t is the target point, which is the point on the straight line with a foresight distance h from the projection point of the lateral error. Then:

[0125] Lateral error x d The calculation formula is:

[0126]

[0127] Thus, the desired angle of the unmanned ship can be determined The expected forward speed u of the unmanned ship is d Set by the user to u d :

[0128]

[0129] Finally, the expected forward speed u d and the desired angle (also called the desired heading) Constitute the expected output at each moment, such as the expected forward speed u at time k+1 d (k+1) and the expected angle at time k+1 Constitute the expected output at time k+1 The expected output g at several moments d (k+1) constitutes the expected output set G d (k+1):

[0130]

[0131] S350: Based on the performance function, the expected output, and the prediction model, determine the optimal solution of the decision variable, and select the control amount of the thruster module corresponding to the current moment as the target control amount from the optimal solution of the decision variable.

[0132] Optionally, the performance function J(k) is used to analyze the degree of agreement between the expected output and the predicted variable (predicted output). The formula is as follows:

[0133]

[0134]

[0135] Among them, the coefficient matrix K g It is composed of pre-set weight coefficients, and the weight k at the i-th moment is gThe larger (i) is, the higher the proportion of the error calculated at the i-th moment in J(k), and usually all of them can take the same value. Specifically:

[0136]

[0137] Therefore, the following optimization problem can be established and solved in model predictive control:

[0138]

[0139] Among them, f min 、f max It is the minimum thrust and maximum thrust that the current first thruster and second thruster can actually achieve.

[0140] In the embodiment of the present application, after solving the above optimization problem, it is assumed that the optimal solution F of the decision variable F(k) is obtained. * (k) is:

[0141]

[0142] Among them, f * (k) represents the optimal solution value at time k. The control amount of the thruster module corresponding to the current time k is selected from the optimal solution of the decision variable, that is, f * (k) As the target control quantity, the target control quantity is the thrust of the first thruster in the target control quantity, is the thrust of the second thruster in the target control quantity.

[0143] In one embodiment, determining the disturbance compensation output according to the disturbance estimation value in step S400 includes steps S410-S430:

[0144] S410: Obtain the moment of inertia of the unmanned ship, calculate the ratio of the moment of inertia to the distance, determine a first product of half the mass of the unmanned ship and the first estimated value, and a second product of the ratio and the second estimated value.

[0145] Optionally, obtain the moment of inertia I of the unmanned ship and calculate the ratio of the moment of inertia I to the distance d Determine the first product of half the mass of the unmanned ship and the first estimate and the second product of the ratio and the second estimate

[0146] S420: Determine a first compensation output according to the inverse of a difference between the first product and the second product, and determine a second compensation output according to the inverse of a sum of the first product and the second product.

[0147] Specifically, in combination with the above formula (13), the first compensation output f is determined according to the inverse of the difference between the first product and the second product. lb , determine the second compensation output f according to the inverse of the sum of the first product and the second product rb .

[0148] S430: Obtain a disturbance compensation output according to the first compensation output and the second compensation output.

[0149] Optionally, the first compensation output f lb , the second compensation output f rb , constituting the disturbance compensation output

[0150] In one embodiment, determining the target control output of the thruster module according to the target control variable and the disturbance compensation output in step S400 includes step S440:

[0151] S440: Calculate the sum of the target control amount and the disturbance compensation output to obtain the target control output of the thruster module.

[0152] Specifically, the target control output f'(k) is calculated as:

[0153] f'(k)=f * (k)+f b (twenty one)

[0154] Therefore, the target control output f'(k) at time k is finally obtained, which includes the target thrust corresponding to the first and second thrusters. It can be understood that at the next time k+1, the target control output corresponding to each time can be obtained by recalculating based on the above principle.

[0155] In an embodiment of the present application, in order to improve the stability and accuracy of the speed control of the nonlinear dynamic system of an unmanned ship without affecting the real-time performance of the control, an RBF neural network is constructed as a disturbance observer to approximate the unmanned ship model, determine the disturbance estimate value, and thus perform disturbance compensation. Compared with other schemes that directly simplify the system model to obtain a linearized model, more accurate unmanned ship speed control is achieved, while avoiding the complicated parameter adjustment in the anti-disturbance control; the problem is converted into an accurately known linear model, avoiding the model error caused by approximation, and using the linear model predictive control to establish a quadratic optimization problem, which can be quickly solved to ensure stability while improving the solution speed without affecting real-time performance, and finally combined with the disturbance compensation output to make up for the error caused by model simplification, thereby ultimately determining an accurate and effective target control output.

[0156] In addition, compared with the related technology that uses the distance formula from a point to a straight line to calculate the lateral error, the lateral error obtained thereby has no positive or negative distinction, and usually requires a layer of human direction judgment. In the line of sight method, this application solves and calculates the lateral error based on the method of the local coordinate system. By calculating the direction angle of the straight line, the lateral error is calculated using the direction angle of the straight line. The lateral error calculation result obtained at this time has a direction according to the direction angle of the straight line and the definition of the coordinate system, so the expected heading angle of the unmanned ship can be directly calculated using the formula.

[0157] Reference Figure 4 , shows a structural block diagram of an unmanned ship prediction tracking control device based on disturbance compensation according to an embodiment of the present application, the device may include:

[0158] a first determining module, configured to determine an unmanned ship model and a thrust value of a propeller module of the unmanned ship, and determine a disturbance value of the propeller module according to the unmanned ship model and the thrust value of the propeller module;

[0159] A first prediction module is configured to predict a disturbance estimation value of the propulsion module based on a radial basis function neural network, a thrust value of the propulsion module, and a disturbance value;

[0160] A second prediction module is used to simplify the unmanned ship model into a linear model based on the disturbance estimation value, and predict the target control amount of the propulsion module according to the linear model and the thrust value of the propulsion module;

[0161] The second determination module is used to determine the disturbance compensation output according to the disturbance estimation value, and to determine the target control output of the thruster module according to the target control amount and the disturbance compensation output.

[0162] The functions of each module in the device of the embodiment of the present application can be found in the corresponding description in the above method and will not be repeated here.

[0163] Reference Figure 5 , shows a block diagram of an electronic device according to an embodiment of the present application. The electronic device includes a memory 310 and a processor 320. The memory 310 stores instructions executable on the processor 320, which loads and executes the instructions to implement the predictive tracking control method for an unmanned vessel based on disturbance compensation in the above embodiment. The number of the memory 310 and the processor 320 can be one or more.

[0164] In one embodiment, the electronic device further includes a communication interface 330 for communicating with external devices and performing data exchange transmission. If the memory 310, the processor 320, and the communication interface 330 are implemented independently, the memory 310, the processor 320, and the communication interface 330 can be interconnected through a bus and complete communication with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0165] Optionally, in a specific implementation, if the memory 310, the processor 320 and the communication interface 330 are integrated on a chip, the memory 310, the processor 320 and the communication interface 330 can communicate with each other through an internal interface.

[0166] An embodiment of the present application provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the unmanned ship predictive tracking control method based on disturbance compensation provided in the above embodiment.

[0167] An embodiment of the present application also provides a chip, which includes a processor for calling and executing instructions stored in the memory from the memory, so that a communication device equipped with the chip executes the method provided in the embodiment of the present application.

[0168] An embodiment of the present application also provides a chip, including: an input interface, an output interface, a processor and a memory. The input interface, the output interface, the processor and the memory are connected through an internal connection path. The processor is used to execute the code in the memory. When the code is executed, the processor is used to execute the method provided in the embodiment of the application.

[0169] It should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. It is worth noting that the processor may be a processor that supports the advanced reduced instruction set machine (ARM) architecture.

[0170] Furthermore, optionally, the above-mentioned memory may include read-only memory and random access memory, and may also include non-volatile random access memory. The memory may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory may include read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) or flash memory. The volatile memory may include random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available. For example, static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct rambus RAM (DR RAM).

[0171] In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0172] In the description of this specification, the reference terms "one embodiment," "some embodiments," "examples," "specific examples," or "some examples" mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials, or characteristics described may be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and integrate different embodiments or examples described in this specification, as well as features of different embodiments or examples, unless they are mutually inconsistent.

[0173] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of technical features indicated. Thus, features specified as "first" or "second" may explicitly or implicitly include at least one of such features. In the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0174] Any process or method description in a flow chart or otherwise described herein can be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process. The scope of the preferred embodiments of the present application includes additional implementations in which the functions may be performed in a different order than shown or discussed, including in a substantially simultaneous manner or in a reverse order depending on the functions involved.

[0175] The logic and / or steps represented in the flowchart or otherwise described herein may be considered, for example, as an ordered list of executable instructions for implementing logical functions, and may be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device).

[0176] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above embodiment method can be completed by instructing the relevant hardware through a program, which can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0177] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the aforementioned integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium. The storage medium may be a read-only memory, a magnetic disk, or an optical disk, etc.

[0178] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included within the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A predictive tracking control method for an unmanned ship based on disturbance compensation, characterized in that: include: Determining an unmanned ship model and a thrust value of a propulsion module of the unmanned ship, and determining a disturbance value of the propulsion module based on the unmanned ship model and the thrust value of the propulsion module; According to the radial basis function neural network, the thrust value of the propeller module and the disturbance value, the disturbance estimation value of the propeller module is predicted; Based on the disturbance estimation value, the unmanned ship model is simplified into a linear model. According to the linear model and the thrust value of the propeller module, the target control variable of the propeller module is predicted. The disturbance compensation output is determined based on the disturbance estimation value, and the target control output of the thruster module is determined based on the target control amount and the disturbance compensation output.

2. The unmanned vessel predictive tracking control method based on disturbance compensation according to claim 1, characterized in that: Determining the unmanned ship model includes: A kinematic model is constructed based on the position of the unmanned ship in the plane coordinate system, the derivative velocity of the position, the rotation velocity, the forward velocity of the unmanned ship along one coordinate axis and the lateral velocity along another coordinate axis in the unmanned ship coordinate system, and the angle between the plane coordinate system and the unmanned ship coordinate system; Constructing a dynamic model based on the spacing between the first propeller and the second propeller in the propulsion module, the mass of the unmanned ship, the differential of the forward speed of the unmanned ship along one of the coordinate axes in the unmanned ship coordinate system, the differential of the rotational speed, the thrust of the first propeller, and the thrust of the second propeller; Among them, the unmanned ship model includes a kinematic model and a dynamic model.

3. The unmanned vessel predictive tracking control method based on disturbance compensation according to claim 2, characterized in that: The thrust value of the thruster module includes the current thrust value of the first thruster and the current thrust value of the second thruster; Calculating the disturbance value of the propeller module according to the unmanned ship model and the thrust value of the propeller module includes: Obtain the actual values ​​of the current derivative speed and angle, and combine them with the kinematic model to calculate the current forward speed and the differential of the current forward speed; The actual values ​​of the rotation speed and the mass of the unmanned ship are obtained, and combined with the dynamic model to determine the disturbance value of the propulsion module. The disturbance value of the propulsion module includes the resistance encountered by the unmanned ship in the direction of the forward speed during movement and the torque of the resistance acting on the center of gravity of the unmanned ship.

4. The unmanned vessel predictive tracking control method based on disturbance compensation according to claim 3, characterized in that: The step of predicting the disturbance estimation value of the propeller module based on the radial basis function neural network, the thrust value of the propeller module, and the disturbance value includes: Constructing an input vector of the input layer of the radial basis function neural network based on the current thrust value of the first propeller, the current thrust value of the second propeller, the angle, the actual value of the rotation speed, and the current forward speed, and using the disturbance value of the propeller module as the output vector; Determine an output vector of a hidden layer of the radial basis function neural network according to the radial basis function neural network, the input vector, and the output vector; According to the output vector of the hidden layer and the weight matrix from the hidden layer to the output layer, the disturbance estimation value of the thruster module is predicted; The disturbance estimation value includes a first estimation value of the resistance experienced by the unmanned ship in the direction of the forward speed during movement and a second estimation value of the moment magnitude of the first estimation value acting on the center of gravity of the unmanned ship.

5. The unmanned vessel predictive tracking control method based on disturbance compensation according to claim 3, characterized in that: The target control amount of the propeller module is predicted based on the linear model and the thrust value of the propeller module, including: Based on the sampling time, the linear model is discretized to determine the discrete state equation; Determine a prediction variable based on the forward speed, angle, and prediction step size, and determine a decision variable based on the control step size. The decision variable is a set of control variables of the thruster module at the current moment and several moments in the future. Each control variable includes the thrust of the first thruster and the thrust of the second thruster. Determine the prediction model based on the discrete state equation, prediction variables and decision variables; Determine the expected forward speed and expected angle of the unmanned vessel, and determine the expected output based on the expected forward speed, expected angle, and predicted step size; Based on the performance function, expected output and prediction model, the optimal solution of the decision variable is determined, and the control amount of the thruster module corresponding to the current moment is selected as the target control amount from the optimal solution of the decision variable.

6. The unmanned vessel predictive tracking control method based on disturbance compensation according to claim 5, characterized in that: Determining the disturbance compensation output according to the disturbance estimation value includes: Obtaining a moment of inertia of the unmanned vessel, calculating a ratio of the moment of inertia to the distance, determining a first product of half the mass of the unmanned vessel and a first estimated value, and a second product of the ratio and a second estimated value; Determining a first compensation output according to the inverse of a difference between the first product and the second product, and determining a second compensation output according to the inverse of a sum of the first product and the second product; A disturbance compensation output is obtained according to the first compensation output and the second compensation output.

7. The unmanned vessel predictive tracking control method based on disturbance compensation according to claim 6, characterized in that: Determining the target control output of the thruster module according to the target control amount and the disturbance compensation output includes: The target control quantity and the sum of the disturbance compensation output are calculated to obtain the target control output of the thruster module.

8. An unmanned ship prediction tracking control device based on disturbance compensation, characterized in that: include: a first determining module, configured to determine an unmanned ship model and a thrust value of a propeller module of the unmanned ship, and determine a disturbance value of the propeller module according to the unmanned ship model and the thrust value of the propeller module; A first prediction module is configured to predict a disturbance estimation value of the propulsion module based on a radial basis function neural network, a thrust value of the propulsion module, and a disturbance value; A second prediction module is used to simplify the unmanned ship model into a linear model based on the disturbance estimation value, and predict the target control amount of the propulsion module according to the linear model and the thrust value of the propulsion module; The second determination module is used to determine the disturbance compensation output according to the disturbance estimation value, and to determine the target control output of the thruster module according to the target control amount and the disturbance compensation output.

9. An electronic device, characterized in that: include: A processor and a memory, wherein the memory stores instructions, and the instructions are loaded and executed by the processor to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, and when the computer program is executed, the method according to any one of claims 1 to 7 is implemented.

Images