A method for very short-term forecasting of ship tracks under the influence of ocean currents

Abstract

This invention discloses a method for extremely short-term ship track forecasting under the influence of ocean currents, comprising: conducting experiments; data acquisition; data storage; redundant data removal; removing obviously unreasonable data and smoothing the remaining data after removing unreasonable data; identifying the parameters of the maneuvering model using the least squares method to obtain a three-degree-of-freedom kinematic model; reading ocean current information within the ship's navigation area; using a ship speed characteristic distribution algorithm and a track center point reconstruction method to perform a weighted average of the ocean current speed information in the navigation area to obtain accurate ocean current information; and forecasting the ship's extremely short-term track. This invention can establish an extremely short-term forecasting model based on ship track data. Based on the ship's turning track and speed distribution characteristics, it extracts the ocean current direction and speed of the turning test, and integrates ocean current information and ship response characteristics into the ship's extremely short-term forecasting model to achieve segmented extremely short-term ship track forecasting, thereby improving the accuracy of ship track forecasting and the safety of navigation.

Description

Technical Field

[0001] This invention belongs to the field of waterway transportation, specifically relating to a method for very short-term forecasting of ship tracks under the influence of ocean currents. Background Technology

[0002] With economic development and the continuous growth in the number of merchant ships worldwide, ship maneuverability is closely related to ship navigation safety and energy conservation, making ship maneuverability an important area of ​​ship safety and energy conservation control.

[0003] Ship handling performance is typically evaluated through handling motion tests, which can be categorized into three types: full-ship tests, model ship tests, and numerical simulation tests. Full-ship tests generally involve conducting Z-turn tests, swerve tests, and stop tests on a full-scale ship in open, calm waters. By analyzing the ship's trajectory and handling commands, parameters such as turning diameter, maneuvering index, and stopping distance are obtained. Model ship tests generally involve conducting static and dynamic planar motion mechanism tests on a scaled-down model in a shipyard environment to obtain ship handling parameters. However, the actual navigation environment of ships is difficult to predict in current technologies, and conducting full-ship tests requires very high experimental conditions and costs. Summary of the Invention

[0004] The purpose of this invention is to provide a method for very short-term forecasting of ship tracks under the influence of ocean currents. This method can establish a very short-term forecasting model based on ship track data, extract the direction and speed of ocean currents during turning tests based on the characteristics of ship turning tracks and speed distribution, and integrate ocean current information and ship response characteristics into the ship very short-term forecasting model to achieve segmented very short-term forecasting of ship tracks, thereby improving the accuracy of ship track forecasting and the safety of navigation.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is: a method for very short-term forecasting of ship tracks under the influence of ocean currents, comprising the following steps:

[0006] S1. Conduct ship maneuverability tests in ocean current conditions, including Z-shaped and turning tests in the test area under common navigation conditions.

[0007] S2. Collect navigation information during ship navigation; navigation information includes ocean current information in the ship's navigation area, ship's speed relative to land, ship's own positioning, navigation attitude data, navigation scenario information, and corresponding ship control rudder angle and main engine speed information;

[0008] S3, Store navigation information;

[0009] S4. Remove redundant data and data unrelated to Z-type test and turning test from the stored navigation information, retain the Z-type test data and turning test data and group them;

[0010] S5. Remove obviously unreasonable data from the Z-type test data and the turning test data, and smooth the remaining data after removing unreasonable data. Use the least squares method to identify the maneuvering model parameters to obtain the three-degree-of-freedom kinematic models of the Z-type test data and the turning test data respectively. Read the ocean current information in the ship's navigation area, and use the ship speed characteristic distribution algorithm and the track center point restoration method to perform a weighted average of the ocean current speed information in the navigation area to obtain accurate ocean current information.

[0011] S6. Make extremely short-term forecasts of ship tracks based on accurate ocean current information.

[0012] In S5, a three-degree-of-freedom kinematic model is used, combined with ocean current information and the ship's speed relative to the ground within the navigation area, to obtain a very short-term prediction model for the ship's trajectory, which predicts the navigation trajectory.

[0013] The navigation scenario information in S2 includes the ship's tonnage, main engine speed, speed, heading, latitude and longitude, and the current water flow speed and direction in the current sea area.

[0014] S3 specifically involves classifying and storing navigation information to generate a ship maneuverability test database.

[0015] Before performing S4, redundant data and data irrelevant to the Z-type and rotational tests are pre-selected. The pre-selection steps are as follows:

[0016] (1) Screening and obtaining Z-type test data and rotation test data;

[0017] (2) Remove duplicate samples from the sampling of Z-type test data and slewing test data respectively, so that there are identical ship state data in adjacent time periods;

[0018] (3) Remove abnormal data from the Z-type test and the slewing test that exceed the normal values, such as excessive changes in rudder angle, excessive changes in rotation speed, excessive changes in speed, and points where the position jumps.

[0019] (4) After removing abnormal data, use linear interpolation to fill in the abnormal data on the remaining data to make the data smoother.

[0020] The method for obtaining the three-degree-of-freedom kinematic model of Z-type experimental data is as follows:

[0021] The first-order nonlinear response model in equation (1) is selected as the motion model of the ship:

[0022]

[0023] Where δ is the rudder angle, r is the pitching angular velocity, K and T are dimensionless coefficients after simplification of the hydrodynamic derivatives, and a is the coefficient of the nonlinear term.r To reduce the rudder angle, the ship's heading angle ψ and the angular velocity of the roll r exist. The correspondence;

[0024] The ship's heading angle, rudder angle, and rudder angle command were extracted from the Z-type test data, and the ship model parameters were extracted using the least squares method.

[0025] The method for obtaining the three-degree-of-freedom kinematic model of the slewing test data is as follows:

[0026] Using a ship speed characteristic distribution algorithm, the ship's speed and heading angle are extracted from the turning test data, and the maximum speed V is extracted respectively. max With minimum speed V min And their corresponding ship heading angles θ1 and θ2, respectively. If θ1 > θ2, then the current water flow direction γ in the current area is:

[0027] γ=0.5(θ1+θ2)+90° (2)

[0028] If θ1≤θ2, then we have:

[0029] γ=0.5(θ1+θ2)-90° (3)

[0030] Among them, the speed of the ocean current V y =0.5cos(θ1-γ)(V max -V min );

[0031] Using the center point restoration method, since the ship's center is a fixed point when turning in still water, the ship's center of rotation moves with the current under the influence of ocean currents. The speed at which the center of rotation moves is the speed of the ocean current in the current area. The track points between the starting point and the ending point of the turning track are {p1, p2, ..., p...}. n}, corresponding to the heading directions {θ1, θ2, ..., θ n When the ship rotates clockwise, the coordinates of the rotation center (x) are... c y c The calculation formula is:

[0032]

[0033] When a ship rotates counterclockwise, the formula for calculating the coordinates of the center of rotation is:

[0034]

[0035] Where R is the radius of gyration in still water;

[0036] The coordinates of the center of rotation are obtained as C(x). c y cAfter that, for the track points {p1, p2, ..., p} n The calculated coordinates of the waypoints are {c1, c2, ..., c}. n}; Fitting the coordinates of the track points yields the line l connecting the center points, with the starting point c1 and ending point c of the line l. n The direction of the line is used to obtain the current water flow direction γ in the region; the fitted straight line l starts at c1 and ends at c1. n distance d l The time t of the ship's turning motion is obtained l The water flows forward a distance d; at this time, the water flow v c Size is v c =d l / (t l ), t l This represents the total time for the selected slewing test data segment;

[0037] In the ship speed characteristic distribution algorithm, the deviation between the maximum ground speed heading and the estimated ocean current direction is caused by interference from uncertain factors during actual navigation; therefore, the resulting deviation is the first deviation, e1. Similarly, in the center point restoration method, the average error of the straight track from the restored actual center point to the fitted center point is also caused by interference; therefore, the resulting deviation is the second deviation, e2. The weighted extracted ocean current speed V... c With direction θ c for:

[0038]

[0039] Among them, V e θ represents the ocean current velocity obtained through a ship speed characteristic distribution algorithm. e V is the direction obtained through the ship speed characteristic distribution algorithm. m θ is the ocean current velocity obtained by the center point reduction method. m The direction is obtained by the center point restoration method;

[0040] Furthermore, considering the possibility that the large deviations in the two prediction methods could lead to significant errors in the final extracted ocean currents, a deviation acceptance range e is set for each method. 1max and e 2max When the deviation of one of the methods exceeds its acceptable range, the estimated ocean current velocity and direction of that method are disregarded, and V is obtained. c Represented as:

[0041]

[0042] S6 specifically refers to:

[0043] In short-run uniform flow, the velocity V of the water flow c With the direction γ remaining unchanged, the ship's speed relative to the water Vs Angle β with the bow s This is caused by the hydrodynamic influence of the ship's relative motion with the ocean current. In a uniform flow field, its magnitude is only related to the rudder angle. The value of the nonlinear term α in the ship's three-degree-of-freedom model can be regarded as a constant value, and the ocean current velocity is also a constant value. Therefore, by combining the three-degree-of-freedom kinematic model with Z-type test data and turning test data and ocean current information, a very short-term prediction of the ship's trajectory is made. The prediction model is as follows:

[0044]

[0045] A computer device is also provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, performs the steps of the method as described in any of the preceding claims.

[0046] A computer-readable storage medium is also provided, on which a computer program is stored, characterized in that, when the computer program is executed by a processor, it implements the steps of the method as described in any of the preceding claims.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0048] This invention obtains a three-degree-of-freedom kinematic model of a ship by using Z-type test data. It then extracts ship current information by weighting the distribution of turning speed characteristics in ocean currents and the method of restoring the turning center point, and constructs a very short-term ship trajectory forecasting model. By applying ocean current data to the very short-term ship trajectory forecasting, it can quickly and accurately predict ship motion, which is beneficial for intelligent and automated ship control and promotes the development of ship automatic control technology. Attached Figure Description

[0049] Figure 1 This is a flowchart illustrating an embodiment of the present invention;

[0050] Figure 2 This is a schematic diagram of the structure of the device according to an embodiment of the present invention;

[0051] Figure 3 This is a flowchart of the center reduction method for ocean current extraction in an embodiment of the present invention;

[0052] Figure 4 This is a flowchart of the ocean current extraction process for the ship speed feature distribution algorithm in this embodiment of the invention;

[0053] Figure 5 This is a flowchart of ocean current weighted extraction in an embodiment of the present invention;

[0054] Figure 6 This is a schematic diagram of very short-term ship trajectory prediction in an embodiment of the present invention;

[0055] Figure 7This is a schematic diagram illustrating the principle of ship trajectory prediction in an embodiment of the present invention.

[0056] Figure 8 This is a schematic diagram illustrating the application of the short-term ship forecasting model in an embodiment of the present invention;

[0057] In the diagram, 1-compass, 2-log, 3-GPS, GNSS and BD, 4-thruster, 5-servo motor. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0059] This invention provides a device and method for short-term ship track forecasting under the influence of ocean currents. It can establish an extremely short-term forecasting model based on ship track data, extract the direction and speed of the ocean current during the turning test based on the characteristics of the ship's turning track and speed distribution, and integrate ocean current information and ship response characteristics into the ship's extremely short-term forecasting model to achieve segmented extremely short-term ship track forecasting, thereby improving the accuracy of ship track forecasting and the safety of navigation.

[0060] One of them is a short-term forecasting device for ship tracks under the influence of ocean currents: it includes equipment sensors mounted on the ship, as well as data acquisition system, data processing, transmission, storage, analysis, processing and calculation units.

[0061] The ship's sensors include a compass to collect the ship's heading and attitude; a current velocity and direction sensor to collect the ship's position and current information; a GPS or BeiDou positioning module to collect the ship's positioning and speed information; an encoder or speed sensor to collect the main engine speed information; and a rudder angle acquisition module. The data acquisition module collects data from all sensors and transmits it to the model building module via a network. The model building module can establish a very short-term forecast model based on the ship's trajectory data. Based on the ship's turning trajectory and speed distribution characteristics, it extracts the direction and speed of the ocean current during the turning test. It integrates ocean current information and ship response characteristics into the ship's very short-term forecast model to achieve segmented very short-term forecasts of the ship's trajectory, thereby improving the accuracy of ship trajectory forecasts and the safety of navigation.

[0062] Using the above-mentioned device, a method for short-term forecasting of ship tracks under the influence of ocean currents is proposed, including the following steps:

[0063] S1. Experimental Implementation:

[0064] By conducting ship maneuverability tests in ocean current environments, Z-turn and swerve tests were carried out in the test area under common navigation conditions.

[0065] S2, Data Acquisition:

[0066] Various types of information are collected during the ship's navigation through various sensors; these include the ship's own positioning, navigation attitude data, navigation scene information, and corresponding ship control rudder angle and speed information.

[0067] S3, Data Storage:

[0068] Store all types of information collected;

[0069] S4. Redundant Data Removal:

[0070] The collected and stored data are processed to remove redundant data and data that are not relevant to the Z-shaped and slewing tests, and the Z-shaped and slewing data are grouped.

[0071] S5. Data Processing:

[0072] The ship's Z-shaped and turning tests under the influence of ocean currents were extracted, unreasonable data were removed, and the original data were smoothed. The least squares method was used to identify the parameters of the maneuvering model, resulting in a three-degree-of-freedom kinematic model. Ocean current speed information in the navigation area was read. When the ship was equipped with a reliable speed log, the ocean current speed information was obtained directly through the speed log. In the absence of a current measuring instrument, the ocean current was estimated through the ship's turning test. The accurate ocean current information was obtained by weighted averaging the ship's speed characteristic distribution and the track center point reconstruction method.

[0073] S6. Very Short-Term Ship Track Forecast:

[0074] According to the above scheme, a three-degree-of-freedom kinematic model is used, combined with ocean current information and ship speed relative to the ground within the ship's navigation area, to obtain a very short-term prediction model for the ship's trajectory, and to predict the navigation trajectory.

[0075] According to the above scheme, the types of ship tests include Z-turn and turning tests under common ship operating conditions. The ship navigation test data specifically includes: the ship's tonnage, main engine speed, speed, course, latitude and longitude information, and the current water flow speed and direction in the current sea area.

[0076] The corresponding ship control information specifically includes: ship rudder angle and main engine speed input.

[0077] According to the above scheme, S3 will classify and store the collected information to generate a ship maneuverability test database.

[0078] According to the above scheme, S4 specifically includes the pre-elimination of redundant data other than the ship's Z-type and turning tests.

[0079] The Z-type test data extracted according to the above scheme are processed to obtain the ship maneuverability model;

[0080] Firstly, for the Z-type test, a first-order nonlinear response model was selected as the motion model of the ship:

[0081]

[0082] Where δ is the rudder angle, r is the pitching angular velocity, K and T are dimensionless coefficients after simplification of the hydrodynamic derivatives, and a is the coefficient of the nonlinear term. r To reduce the rudder angle, the bow direction ψ and the bow roll angular velocity r exist. For the original data, the ship's heading angle, rudder angle, and rudder angle commands are extracted, and the least squares method is used to extract the ship model parameters. For ships equipped with reliable logs, the ship's current speed and direction information are directly read. For ships lacking logs, a swivel test is used to estimate the ocean current.

[0083] For the turning test data, the ship's speed and heading were extracted and analyzed using speed characteristic distribution and algorithms, and the maximum speed V was extracted respectively. max With minimum speed V min Point speed and heading information θ1 and θ2; if θ1 > θ2, the direction of the ocean current is:

[0084] γ=0.5(θ1+θ2)+90° (2)

[0085] If θ1≤θ2, the direction of the ocean current is:

[0086] γ=0.5(θ1+θ2)-90° (3)

[0087] The speed of the ocean current V y =0.5cos(θ1-γ)(V max -V min ).

[0088] Using the center point restoration method, since the ship's center is a fixed point when turning in still water, it shifts with the current due to the influence of the ocean current. The speed at which the center point moves is the speed of the ocean current in the current area. The track points between the starting and ending points of the turning track are {p1, p2, ..., p...}. n}, corresponding to the heading directions {θ1, θ2, ..., θ n When the ship rotates clockwise, the coordinates of the rotation center (x) are... c y c The calculation formula is:

[0089]

[0090] When a ship rotates counterclockwise, the formula for calculating the coordinates of the center of rotation is:

[0091]

[0092] R is the radius of gyration in still water.

[0093] The coordinates of the center of rotation are obtained as C(x). c y c After that, for the track points {p1, p2, ..., p} n The coordinates of the track points can be calculated as {c1, c2, ..., c}. n By fitting the coordinates of the track points, we can obtain the line l connecting the center points. The starting point c1 and the ending point c of the line l are... n The direction of the line connecting the points gives the current water flow direction γ in the region. The fitted straight line l is connected from its starting point c1 to its ending point c. n distance d l The time t of the ship's turning motion is obtained l The water flows forward a distance d. At this time, the water flow v c Size is v c =d l / (t l ), t l This represents the total time for the selected slewing test data segment.

[0094] After the ship enters the turning state, the turning part of the ship's trajectory is selected. Based on the turning radius of the ship in still water and the ship's heading at the current moment, the turning center of the ship can be obtained. The turning center point of the ship is fitted, and the distance of the fitted straight line is calculated. Then, based on the distance of the center point and the turning time, the average flow velocity in the turning area can be obtained. The direction of the fitted straight line can be used to obtain the direction of the water flow in the test area.

[0095] The deviation between the maximum ground velocity heading and the estimated ocean current direction in the maximum-minimum velocity difference method is caused by interference from uncertainties in actual navigation; let's assume the deviation is e1. Similarly, the average error of the straight track from the restored actual center point to the fitted center point in the center point restoration method is also due to interference; let's assume the deviation is e2. Since the deviation values ​​are on the same order of magnitude, no normalization is performed. Let's assume the ocean current velocity and direction obtained in the maximum-minimum velocity difference method are V. e and θ e The ocean current velocity and direction obtained in the center point restoration method are V. m and θ m The weighted extracted ocean current velocity V c With direction θ c for:

[0096]

[0097] Furthermore, considering the possibility that the large deviations in the two prediction methods could lead to significant errors in the final extracted ocean currents, a deviation acceptance range e is set for each method. 1max and e 2max When the deviation of one of the methods exceeds its acceptable range, the estimated ocean current velocity and direction of that method are disregarded. Therefore, V c It can be represented as:

[0098]

[0099] According to the above scheme, the ship maneuverability model and accurate information on the ship's current speed and direction in the current sea area can be obtained. Specifically, S6 is:

[0100] In short-run uniform flow, the velocity V of the water flow c The direction γ remains unchanged, and the angle β between the ship's velocity relative to the water and its heading remains constant. s This phenomenon arises from the hydrodynamic influence of the ship's relative motion with the ocean current. In a uniform flow field, its magnitude is usually only related to the rudder angle. The value of the nonlinear term α in the ship's three-degree-of-freedom model can be considered constant, and the ocean current velocity is also constant. Therefore, the ship's trajectory can be predicted in the very short term by combining the ship's maneuverability model with ocean current information. The prediction model is as follows:

[0101]

[0102] Ocean current speed information V c γ can be calculated from S5.

[0103] The beneficial effects of this invention are as follows: By obtaining a three-degree-of-freedom kinematic model of a ship using Z-type test data, and by combining the ship's turning speed characteristics distribution in the ocean current with the method of restoring the turning center point, a ship track very short-term forecast model is constructed. Applying ocean current data to the very short-term forecast of ship tracks can quickly and accurately predict ship motion, thus promoting the development of ship automatic control technology for intelligent and automated ship control.

[0104] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for very short-term forecasting of ship tracks under the influence of ocean currents, characterized in that, Includes the following steps: S1. Conduct ship maneuverability tests in ocean current conditions, including Z-shaped and turning tests in the test area under common navigation conditions. S2. Collect navigation information during ship navigation; navigation information includes ocean current information in the ship's navigation area, ship's speed relative to land, ship's own positioning, navigation attitude data, navigation scenario information, and corresponding ship control rudder angle and main engine speed information; S3, Store navigation information; S4. Remove redundant data and data unrelated to Z-type test and turning test from the stored navigation information, retain the Z-type test data and turning test data and group them; S5. Remove obviously unreasonable data from the Z-type test data and the turning test data, and smooth the remaining data after removing unreasonable data. Use the least squares method to identify the maneuvering model parameters to obtain the three-degree-of-freedom kinematic models of the Z-type test data and the turning test data respectively. Read the ocean current information in the ship's navigation area, and use the ship speed characteristic distribution algorithm and the track center point restoration method to perform a weighted average of the ocean current speed information in the navigation area to obtain accurate ocean current information. S6. Make extremely short-term forecasts of ship tracks based on accurate ocean current information; The method for obtaining the three-degree-of-freedom kinematic model of the Z-type test data is as follows: The first-order nonlinear response model in equation (1) is selected as the motion model of the ship: （1） in, As the rudder angle, The initial rocking angular velocity, T represents the dimensionless coefficient after simplification of the hydrodynamic derivative. For nonlinear term coefficients, To reduce the rudder angle, the ship's heading angle With the initial angular velocity exist The correspondence; The ship's heading angle, rudder angle, and rudder angle command were extracted from the Z-type test data, and the ship model parameters were extracted using the least squares method. The method for obtaining the three-degree-of-freedom kinematic model of the slewing test data is as follows: Using a ship speed characteristic distribution algorithm, the ship's speed and heading angle are extracted from the turning test data, and the maximum speed is extracted separately. with minimum speed and their corresponding ship heading angles and ,like Then there is the current direction of water flow in the area. : （2） like Then we have: （3） Among them, the speed of ocean currents ; Using the center-point restoration method, since the ship's center is a fixed point when turning in still water, it shifts with the current due to the influence of ocean currents. The speed at which the center of rotation moves is the speed of the ocean current in the current area. The track point between the starting and ending points of the turning track is... Corresponding heading direction When the ship rotates clockwise, the coordinates of the rotation center are... The calculation formula is: （4） When a ship rotates counterclockwise, the formula for calculating the coordinates of the center of rotation is: （5） in, Let be the radius of gyration in still water; The coordinates of the center of rotation are obtained as follows: Afterwards, the waypoints The calculated coordinates of the waypoint are The center point line is obtained by fitting the coordinates of the track points. ,straight line starting point and the finish line The direction of the line indicates the direction of water flow in the current area. Fitted straight line starting point and the finish line distance Obtain the ship's turning time Internal water flow distance At this time, the water flow Size is , This represents the total time for the selected slewing test data segment; The deviation between the maximum ground speed heading and the predicted ocean current direction in the ship speed characteristic distribution algorithm is caused by the interference of uncertain factors in actual navigation; therefore, the obtained deviation is the first deviation amount. In the center point restoration method, the average error of the straight track from the restored actual center point to the fitted center point is also caused by interference; therefore, the resulting deviation is the second deviation. The weighted extracted ocean current velocity With direction for: （6） in, The ocean current speed is obtained through a ship speed characteristic distribution algorithm. The direction obtained through the ship speed characteristic distribution algorithm, The ocean current velocity is obtained using the center point reduction method. The direction is obtained by the center point restoration method; Furthermore, considering the possibility that the large deviations in the two prediction methods could lead to significant errors in the final extracted ocean currents, acceptable deviation ranges were set for both methods. and When the deviation of one of the methods exceeds its acceptable range, the ocean current velocity and direction predicted by that method are disregarded. Represented as: （7）； Specifically, S6 is: In short-term uniform flow, the velocity of the water flow and direction The ship's speed relative to the water remains unchanged. Angle with bow This is caused by the hydrodynamic influence of the ship's relative motion with the ocean current. In a uniform flow field, its magnitude depends only on the rudder angle. The nonlinear term in the ship's three-degree-of-freedom model... The value can be considered constant, and the ocean current speed is also constant. Therefore, by combining the three-degree-of-freedom kinematic model of Z-type test data and slewing test data with ocean current information, a very short-term prediction of the ship's trajectory is made. The prediction model is as follows: （8）。 2. The method for very short-term ship tracking forecasting under the influence of ocean currents according to claim 1, characterized in that, In S5, a three-degree-of-freedom kinematic model is used, combined with ocean current information and the ship's speed relative to the ground within the navigation area, to obtain a very short-term prediction model for the ship's trajectory, which predicts the navigation trajectory.

3. The method for very short-term ship track forecasting under the influence of ocean currents according to claim 1, characterized in that, The navigation scenario information in S2 includes the ship's tonnage, main engine speed, speed, heading, latitude and longitude, and the current water flow speed and direction in the current sea area.

4. The method for very short-term ship track forecasting under the influence of ocean currents according to claim 1, characterized in that, S3 specifically involves classifying and storing navigation information to generate a ship maneuverability test database.

5. The method for very short-term ship track forecasting under the influence of ocean currents according to claim 1, characterized in that, Before performing S4, redundant data and data irrelevant to the Z-type and rotational tests are pre-selected. The pre-selection steps are as follows: (1) Screening to obtain Z-type test data and rotation test data; (2) Remove duplicate samples from the sampling of Z-type test data and slewing test data respectively, so that there are identical ship state data in adjacent time periods; (3) Remove abnormal data from the Z-type test and the slewing test that exceed the normal values, such as excessive changes in rudder angle, excessive changes in rotation speed, excessive changes in speed, and points where the position jumps. (4) After removing abnormal data, use linear interpolation to fill in the abnormal data on the remaining data to make the data smoother.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1-5.

Images