Hull navigation control method, device and system

Abstract

The invention discloses a ship navigation control method, device and system, and belongs to the technical field of ship navigation control, and the method comprises the steps: obtaining the structure data and power data of a ship, and the meteorological data and hydrological data in a ship navigation environment; inputting all the obtained data into a trained ship body motion control model to obtain ship body navigation control parameters; performing sailing control on the ship body according to the sailing control parameters; according to the method, besides the influence of the structure of the ship body on the motion, environmental factors are taken into consideration, and the influence of meteorological conditions and hydrological conditions on the motion of the ship body is added, so that the control on the ship body is more accurate, and the track error of the ship body in the sailing process is reduced; by introducing a visual digital twinborn platform, a prediction and compensation mode is adopted to ensure that a ship can accurately sail according to a planned track in an actual sailing process to the maximum extent.

Description

Technical Field

[0001] The invention relates to a ship navigation control method, device and system, belonging to the technical field of ship navigation control. Background Art

[0002] Traditional ships are used to carry people and cargo, and their navigation is usually controlled by crew members. However, with the increasing demand, their application scenarios are also constantly enriched. They are no longer limited to carrying people and cargo, but also include measurement, exploration, remote rescue, material delivery, and applications in national defense. Their navigation waters are not necessarily familiar waters, and they are more likely to face unknown waters, extreme weather conditions, and long-term endurance. In this case, it is no longer suitable for crew members to drive the ship, and unmanned ship navigation control came into being in this context.

[0003] Unmanned navigation control of ships points the way to new areas of ship applications. The key to unmanned navigation control is the ability to automatically plan paths, and the ship can accurately navigate according to the planned track. However, the automatic planning of routes that can be performed at present can be performed well for simple tracks. Once complex tracks and severe weather conditions are involved, the actual track of the ship will have a large deviation from the planned track. The reason is that in the process of executing the planned track, the ship's various status information is usually collected in real time, and the deviation between the current collected information and the planned track is corrected in real time. However, the ship is affected by inertia during navigation, and its real-time response ability is poor. Therefore, even if a deviation from the planned track is found and corrected, it is difficult to complete the deviation correction immediately. Other deviations will also be introduced during the correction process, which will eventually lead to a large deviation between the planned track and the actual track. Summary of the invention

[0004] The object of the present invention is to provide a ship navigation control method, device and system to solve the problem of large track error existing in the prior art.

[0005] To achieve the above objectives, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a method for controlling a ship's navigation, comprising: Obtain the structural and dynamic data of the ship, as well as the meteorological and hydrological data of the ship's navigation environment; Input all acquired data into the trained ship motion control model to obtain the navigation control parameters of the ship; Performing navigation control on the hull according to the navigation control parameters; The hull motion control model is a deep learning model constructed by taking the structural data and power data of the ship as well as the meteorological data and hydrological data in the ship's navigation environment as independent variables and the navigation control parameters of the hull as dependent variables.

[0006] Further, the structural data includes the hull structure, hull length, hull width and linear characteristics, the power data includes power control data, rudder angle control data and hull weight, the meteorological data includes ambient wind speed, ambient wind direction and ambient temperature, and the hydrological data includes sea state data, water depth, water flow speed and water flow direction; The navigation control parameters include the main engine rotation speed and the rudder angle value.

[0007] Furthermore, the hull motion control model is trained by the following method: The independent variables under historical time are obtained as training parameters, and the training parameters are converted into multidimensional arrays as inputs of the hull motion control model. The difference between the actual track and the planned track of the hull is used as a penalty function to train the hull motion control model.

[0008] Furthermore, when the independent variable is updated, the updated independent variable is used as a new training parameter to retrain the hull motion control model.

[0009] Furthermore, the hull motion control model is a fully connected structure, with multiple fully connected layers, each fully connected layer is provided with multiple nodes, and there are 2 output nodes, and the 2 output nodes correspond to the output main engine speed and the rudder angle value respectively.

[0010] In a second aspect, the present invention provides a ship navigation control device, comprising: The data acquisition module is configured to: acquire structural data and power data of the hull, as well as meteorological data and hydrological data in the hull navigation environment; The navigation control parameter calculation module is configured to: input all acquired data into the trained hull motion control model to obtain the hull's navigation control parameters; A navigation control module is configured to: perform navigation control on the hull according to the navigation control parameters; The hull motion control model is a deep learning model constructed by taking the structural data and power data of the ship as well as the meteorological data and hydrological data in the ship's navigation environment as independent variables and the navigation control parameters of the hull as dependent variables.

[0011] In a third aspect, the present invention provides a ship navigation control system, comprising a visual digital twin platform, a data storage management unit and a ship navigation control unit, wherein the ship navigation control unit is used to execute any one of the methods described in the first aspect; A three-dimensional twin model of the hull is constructed on the visualization digital twin platform, a three-dimensional water twin scene and a three-dimensional space twin scene are built, and a motion control interface of the hull is constructed. The three-dimensional twin model of the hull is placed in the three-dimensional water twin scene. The three-dimensional water twin scene is provided with the hydrological data to simulate the hydrological characteristics of the hull navigation environment, and the three-dimensional space twin scene is provided with the meteorological data to simulate the meteorological characteristics of the hull navigation environment; The motion control interface includes a first control interface affected by the structural data, a second control interface affected by the power data, a third control interface affected by the meteorological data, and a fourth control interface affected by the hydrological data; The data storage management unit is respectively connected to the visual digital twin platform and the hull navigation control unit, and is used to store the structural data and power data of the hull, as well as the meteorological data and hydrological data in the hull navigation environment; The three-dimensional twin model of the hull is jointly controlled by the first control interface, the second control interface, the third control interface and the fourth control interface to perform visualized twin hull navigation control in the three-dimensional water twin scene. In this process, the required navigation control parameters are obtained through the hull navigation control unit, and the navigation control of the hull in the real scene is performed according to the result of the visualized twin hull navigation control.

[0012] Furthermore, it also includes an external data acquisition device, which is communicatively connected to the visual digital twin platform, and the visual digital twin platform can directly obtain data from the external data acquisition device to construct a three-dimensional twin scene.

[0013] Furthermore, the three-dimensional twin model of the hull is constructed using the structural data and power data of the hull.

[0014] Compared with the prior art, the beneficial effects achieved by the present invention are: The present invention provides a method, device and system for controlling the navigation of a ship. In addition to considering the influence of the structure of the ship itself on the movement, environmental factors are also taken into consideration, and the influence of meteorological conditions and hydrological conditions on the movement of the ship is added, so that the control of the ship is more accurate and the track error of the ship during navigation is reduced. By building a digital twin visualization platform, establishing a three-dimensional scene, inserting a twinned three-dimensional model of the hull, and building a ship's motion control interface, the entire process of ship navigation in the three-dimensional scene can be visualized; The training of the hull motion control model and the continuous accumulation of actual navigation data in the subsequent use process can continuously optimize the model and its accuracy; The predicted track is achieved by first simulating in a three-dimensional twin scene (visualized twin hull navigation control), and then the track error generated during the visualized twin hull navigation control process is compensated in the real scene. This prediction and compensation method ensures to the greatest extent that the ship can accurately navigate according to the planned track during actual navigation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a flow chart of a ship navigation control method provided by an embodiment of the present invention; Figure 2 is a schematic diagram of a ship navigation control system provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of the principle of the hull motion control model provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and are not intended to limit the protection scope of the present invention.

[0017] Example 1.

[0018] like Figure 1 As shown, the present invention provides a ship navigation control method, comprising: Obtain the structural and dynamic data of the ship, as well as the meteorological and hydrological data of the ship's navigation environment; Input all acquired data into the trained ship motion control model to obtain the navigation control parameters of the ship; Controlling the hull according to the navigation control parameters; The hull motion control model is a deep learning model constructed by taking the ship's structural data and power data, as well as the meteorological data and hydrological data in the ship's navigation environment as independent variables, and the hull's navigation control parameters as dependent variables.

[0019] In addition to considering the impact of the hull's own structure on the movement, the present invention also takes environmental factors into consideration, adding the impact of meteorological conditions and hydrological conditions on the hull's movement, making the control of the hull more precise and reducing the track error during the hull's navigation process.

[0020] Example 2.

[0021] The present invention provides a ship navigation control method, comprising: S1. Obtain the structural data and power data of the ship, as well as the meteorological data and hydrological data of the ship's navigation environment; S2, input all acquired data into the trained ship motion control model to obtain the navigation control parameters of the ship; S3. Control the navigation of the ship according to the navigation control parameters.

[0022] The hull motion control model is constructed by the following method: Figure 3 As shown in the figure, the ship's structural data and power data as well as the meteorological data and hydrological data in the ship's navigation environment are taken as independent variables (input), and the navigation control parameters of the hull are taken as dependent variables (output); among them, the structural data include the hull structure, hull length, hull width and linear characteristics, the power data include power control data (speed, heading), rudder angle control data (rudder direction) and hull weight, the meteorological data include ambient wind speed, ambient wind direction and ambient temperature, and the hydrological data include sea condition data (wave height), water depth, water flow rate ( Figure 3 Flow velocity in) and water flow direction ( Figure 3 navigation control parameters include main engine speed and rudder angle.

[0023] like Figure 3 As shown in the figure, the hull motion control model is a deep learning model with a fully connected structure. The multiple sets of input parameters at historical moments are converted into a 14×1 multidimensional array structure as the input of the deep learning model. The model sets 3 fully connected layers, and the nodes of each fully connected layer are set to 20. The final output node is 2, which are the main engine speed and rudder angle values. The overall structure of the model is 14×20×20×20×2. The difference between the actual track and the planned track of the hull is used as the penalty function to train the model. The trained deep learning model can generate the main engine speed and rudder angle values ​​as control according to the input parameters such as hull structure, hull length, hull width, hull weight, heading, speed, bow direction, wind direction, wind speed, temperature, wave height, water depth, flow direction, flow velocity, etc., that is, the trained deep learning model is used as the hull motion control model.

[0024] If the model is to be applied to other hulls, the constructed deep learning model can be modified by transfer learning based on the trained hull motion control model. Based on the trained hull motion control model, a high-precision hull motion control model suitable for other ship types can be obtained after simple training.

[0025] For the hull motion control model that has been trained, as the amount of acquired hull navigation data increases, the hull motion control model is continuously optimized and iteratively trained to gradually improve the accuracy of the hull motion control model until the optimized iterative training can no longer continue to improve the control accuracy of the model.

[0026] Example 3.

[0027] The present invention provides a hull navigation control system, comprising a visual digital twin platform, a data storage management unit and a hull navigation control unit, wherein the hull navigation control unit is used to execute the method provided in Example 1 / Example 2.

[0028] The visual digital twin platform uses Django (an advanced Python web framework that can quickly develop secure and maintainable websites) as the framework and three.js (a 3D engine running in the browser) as the rendering engine to improve the realism of the visual model and realize the motion control of the ship model on the digital twin platform.

[0029] A three-dimensional twin model of the hull is constructed on the visual digital twin platform, a three-dimensional water twin scene and a three-dimensional space twin scene are constructed, and a motion control interface of the hull is constructed. The three-dimensional twin model of the hull is placed in the three-dimensional water twin scene. The three-dimensional water twin scene is provided with hydrological data (specifically including the detailed hydrological data listed in Example 2) to simulate the hydrological characteristics of the hull navigation environment, and the three-dimensional space twin scene is provided with meteorological data (specifically including the detailed meteorological data listed in Example 2) to simulate the meteorological characteristics of the hull navigation environment. There are two data pathways, one is the relevant data information already stored in the database (data storage management unit), and the other is the data directly connected to the external data acquisition device.

[0030] In addition to the factors affecting the navigation of the ship itself, its structural characteristics will also affect the movement of the ship. Based on the above reasons, the motion control interface includes a first control interface affected by structural data, a second control interface affected by power data, a third control interface affected by meteorological data, and a fourth control interface affected by hydrological data. Ultimately, the hull motion control based on the joint control of multiple factors is realized, and the visual navigation of the twin hull in the set scenario is realized.

[0031] The data storage management unit is connected to the visualization digital twin platform and the hull navigation control unit respectively, and is used to store the structural data and power data of the hull, as well as the meteorological data and hydrological data in the hull navigation environment.

[0032] Through the first control interface, the second control interface, the third control interface and the fourth control interface, the three-dimensional twin model of the hull is jointly controlled by visual twin hull navigation in the three-dimensional water twin scene. In this process, the required navigation control parameters are obtained by the hull navigation control unit. According to the result of the visualized twin hull navigation control, the hull in the real scene is controlled by navigation, and the track error generated during the visualized twin hull navigation control is compensated.

[0033] Based on the constructed visual digital twin platform and the hull motion control model in the hull navigation control unit, it can realize functions such as automatic trajectory planning, multi-ship formation navigation, dynamic navigation data deduction, and navigation process review.

[0034] Automatic trajectory planning is implemented by preliminarily designing the shortest trajectory and speed that can pass through obstacles based on the target position and discovered obstacles, and then controlling the movement of the 3D twin model of the hull through the hull motion control model on the visual digital twin platform according to the set trajectory and speed. On this basis, the trajectory is continuously corrected to eventually form a planned trajectory with a smooth transition.

[0035] Multi-ship formation navigation is based on the planned track. The relative position relationship between the ships in the formation is set. The main ship sails according to the planned track, and other ships generate navigation tracks and control parameters during the navigation process through the corresponding hull motion control model. The generated track data and control parameters are saved, and the relevant data at any time and position can be loaded at any time for re-analysis as needed.

[0036] Automatic trajectory planning is performed on the visual digital twin platform. In addition to planning the navigation route, the trajectory planning process also combines the meteorological and hydrological information in the digital twin scene, and plans the ship's main engine speed, rudder angle value, etc. based on the hull's motion control model to ensure that the ship can navigate according to the planned route and speed.

[0037] When conducting multi-ship formation navigation on the visual digital twin platform, in addition to considering meteorological information, hydrological information, etc., the mutual influence between ships in the formation is also taken into consideration. The navigation parameters of each ship are set comprehensively based on the formation positions of different ships to ensure that the ships navigate accurately according to the planned track and that the overall formation of the formation is consistent with the preset plan.

[0038] The navigation process is reviewed, the actual track and planned track of the ship are comprehensively compared, the navigation accuracy is evaluated, and the error results are analyzed. The error between the planned track position and the actual track position is used as the penalty function to optimize the iterative training of the ship's hull motion control model to further improve the accuracy of the hull motion control model.

[0039] Compared with the prior art, the advantages of the present invention are: Process visualization: By building a digital twin visualization platform, establishing a three-dimensional scene, inserting a twinned three-dimensional model of the hull, and building a motion control interface for the ship, the full process visualization of the three-dimensional scene can be achieved; Model complexity: In addition to considering the basic conventional hull dynamic factors and structural factors, the built hull motion model also incorporates meteorological conditions and hydrological factors of the navigation waters into the scope of consideration of hull motion. Although the complexity of the model has increased, the hull motion control model has also become more realistic; Iteration of control model: Build a deep learning model pair and perform iterative training on it. Through the iteration of deep learning, a more accurate hull motion control model can be obtained. In the subsequent use process, as the actual navigation data continues to accumulate, the hull motion control model can be iteratively optimized to continuously optimize its accuracy; Traceability of the navigation process: The ship navigates according to the planned route and saves relevant data of the actual navigation process into the database. The navigation process can be reviewed later, the navigation accuracy can be analyzed, and the reasons can be found.

[0040] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0041] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that specifies the functions of a block or multiple blocks.

[0042] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0043] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0044] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A ship navigation control method, characterized in that: include: Obtain the structural and dynamic data of the ship, as well as the meteorological and hydrological data of the ship's navigation environment; Input all acquired data into the trained ship motion control model to obtain the navigation control parameters of the ship; Performing navigation control on the hull according to the navigation control parameters; The hull motion control model is a deep learning model constructed by taking the structural data and power data of the ship as well as the meteorological data and hydrological data in the ship's navigation environment as independent variables and the navigation control parameters of the hull as dependent variables.

2. The ship navigation control method according to claim 1, characterized in that: The structural data includes the hull structure, hull length, hull width and linear characteristics; the power data includes power control data, rudder angle control data and hull weight; the meteorological data includes ambient wind speed, ambient wind direction and ambient temperature; the hydrological data includes sea state data, water depth, water flow speed and water flow direction; The navigation control parameters include the main engine rotation speed and the rudder angle value.

3. The ship navigation control method according to claim 1, characterized in that: The hull motion control model is trained by the following method: The independent variables under historical time are obtained as training parameters, and the training parameters are converted into multidimensional arrays as inputs of the hull motion control model. The difference between the actual track and the planned track of the hull is used as a penalty function to train the hull motion control model.

4. The ship navigation control method according to claim 3, characterized in that: When the independent variable is updated, the updated independent variable is used as a new training parameter to retrain the hull motion control model.

5. The ship navigation control method according to claim 1, characterized in that: The hull motion control model is a fully connected structure, with multiple fully connected layers, each fully connected layer is provided with multiple nodes, and there are 2 output nodes, which correspond to the output main engine speed and the rudder angle value respectively.

6. A ship navigation control device, characterized in that: include: The data acquisition module is configured to: acquire structural data and power data of the hull, as well as meteorological data and hydrological data in the hull navigation environment; The navigation control parameter calculation module is configured to: input all acquired data into the trained hull motion control model to obtain the hull's navigation control parameters; A navigation control module is configured to: perform navigation control on the hull according to the navigation control parameters; The hull motion control model is a deep learning model constructed by taking the structural data and power data of the ship as well as the meteorological data and hydrological data in the ship's navigation environment as independent variables and the navigation control parameters of the hull as dependent variables.

7. A ship navigation control system, characterized in that: It comprises a visualization digital twin platform, a data storage management unit and a hull navigation control unit, wherein the hull navigation control unit is used to execute the method according to any one of claims 1 to 5; A three-dimensional twin model of the hull is constructed on the visualization digital twin platform, a three-dimensional water twin scene and a three-dimensional space twin scene are built, and a motion control interface of the hull is constructed. The three-dimensional twin model of the hull is placed in the three-dimensional water twin scene. The three-dimensional water twin scene is provided with the hydrological data to simulate the hydrological characteristics of the hull navigation environment, and the three-dimensional space twin scene is provided with the meteorological data to simulate the meteorological characteristics of the hull navigation environment; The motion control interface includes a first control interface affected by the structural data, a second control interface affected by the power data, a third control interface affected by the meteorological data, and a fourth control interface affected by the hydrological data; The data storage management unit is respectively connected to the visual digital twin platform and the hull navigation control unit, and is used to store the structural data and power data of the hull, as well as the meteorological data and hydrological data in the hull navigation environment; The three-dimensional twin model of the hull is jointly controlled by visual twin hull navigation in the three-dimensional water twin scene through the first control interface, the second control interface, the third control interface and the fourth control interface. In this process, the required navigation control parameters are obtained through the hull navigation control unit. According to the result of the visual twin hull navigation control, the hull in the real scene is controlled to navigate, and the track error generated during the visual twin hull navigation control is compensated.

8. The ship navigation control system according to claim 7, characterized in that: It also includes an external data acquisition device, which is communicatively connected to the visualization digital twin platform. The visualization digital twin platform can directly obtain data from the external data acquisition device to construct a three-dimensional twin scene.

9. The ship navigation control system according to claim 7, characterized in that: The three-dimensional twin model of the hull is constructed using the structural data and power data of the hull.

Images