Ship berthing manipulation control method, device and equipment and medium
By constructing a three-degree-of-freedom ship maneuvering model and a nonlinear predictive model control, the problem of insufficient dynamic adaptability in multi-ship cooperation was solved, and the efficiency and safety of ship berthing were improved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- WUHAN UNIV OF TECH
- Filing Date
- 2026-01-14
- Publication Date
- 2026-05-05
AI Technical Summary
The existing ship berthing maneuvering control system lacks dynamic adaptability for multi-ship collaboration. The collaborative operation process of multiple entities in the "ship-tug-port" under shore-based control is unclear, and the allocation of roles and the connection of tasks are chaotic, resulting in low berthing efficiency and compromised safety.
A three-degree-of-freedom ship maneuvering model is constructed, in which the controlled ship, the guide ship, and the propulsion ship cooperate. A nonlinear predictive model is used to predict the ship's state and control inputs. The optimal solution is obtained by adaptively adjusting the weight parameters of the cost function to achieve precise control.
It improves the dynamic adaptability and collaborative control efficiency of multi-ship cooperation, and enhances the efficiency and safety of ship berthing.
Smart Images
Figure CN121979207A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of intelligent shipping and ship control technology, and in particular to a ship berthing maneuvering control method, device, equipment and medium. Background Technology
[0002] Port berthing and handling is one of the most dangerous operations in the shipping industry. Due to the complexity of the operation scenario, the numerous interactions among the participants, and the congested navigation environment, it still relies on the traditional "strong man in the ring" model. This results in problems such as outdated work paradigms, low coordination efficiency, and prominent safety hazards in the "ship-port-tug" transportation triangle.
[0003] Existing research has enabled remote monitoring and control of ships using shore-based navigation technology. Collaborative control technology has been adopted to achieve multi-ship cooperation through towing, propulsion, and other modes. However, existing ship berthing and maneuvering control has the following key problems: insufficient dynamic adaptability of multi-ship cooperation; unclear "ship-tug-port" multi-entity collaborative operation process under shore-based navigation control; chaotic role allocation and task connection, resulting in low berthing efficiency; and insufficient model accuracy. The ship motion model does not fully consider interference factors such as port water flow and wind, resulting in berthing trajectory deviation, which in turn affects the safety of ship berthing. Summary of the Invention
[0004] In view of this, it is necessary to provide a method, device, equipment and medium for ship berthing maneuvering control to solve the technical problems of low berthing efficiency and low safety of ships operating in multi-entity collaborative operations.
[0005] To address the aforementioned problems, in a first aspect, the present invention provides a ship berthing maneuvering control method, comprising: A three-degree-of-freedom ship maneuvering model is constructed, in which the controlled ship, the guide ship, and the propulsion ship cooperate. Based on the three-degree-of-freedom ship maneuvering model, the reference path for the ship's berthing, and the current state of the controlled ship, a nonlinear predictive model is used to predict the state of the controlled ship and the control inputs of the guide ship and the propulsion ship, so as to obtain the state of the controlled ship and the control inputs of the guide ship and the propulsion ship at the next moment. Based on the controlled vessel's state at the next moment, the control inputs of the guide vessel and the propeller, and the controlled vessel's state at the current moment, a cost function for nonlinear predictive model control is constructed. With the goal of minimizing the value of the cost function, the weight parameters of the cost function are adaptively adjusted to obtain the optimal solution. Based on the optimal solution, the optimal state of the controlled vessel and the optimal control inputs of the guide vessel and the propeller are obtained at the next moment. The controlled vessel is then controlled based on the optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propeller. The optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propeller are used as the new state of the controlled vessel at the current moment and the control inputs of the guide vessel and the propeller, and this process is iteratively updated to achieve the vessel berthing operation.
[0006] In one possible implementation, the three-degree-of-freedom ship maneuvering model is as follows: , in, For the quality of the controlled vessel, , This is a coefficient related to the ship's hydrodynamics in longitudinal motion. For lateral velocity, This is a coefficient related to the ship's hydrodynamics in lateral motion. For longitudinal velocity, The angular velocity of the bow. The sum of the external moments generated in the longitudinal motion by the forces exerted on the controlled vessel by the guiding vessel and the propulsion vessel. The sum of the external moments generated by the forces exerted on the controlled vessel by the guiding vessel and the propulsion vessel on its lateral motion. The sum of the external torques generated by the forces exerted on the controlled vessel by the guiding vessel and the propulsion vessel on the bow motion. Let be the derivative of the bow angular velocity with respect to time. Let be the derivative of the transverse velocity with respect to time. The derivative of the longitudinal velocity with respect to time. The moment of inertia of the controlled vessel. This is a coefficient related to the ship's hydrodynamics during bow rolling motion.
[0007] In one possible implementation, the state of the controlled vessel includes longitudinal speed, lateral speed, vessel position, heading angular velocity, and bow roll angular velocity, and the control inputs include the thrust and thrust direction of the guide vessel acting on the controlled vessel, and the thrust and thrust direction of the propulsion vessel acting on the controlled vessel.
[0008] In one possible implementation, the cost function for constructing the nonlinear predictive model control based on the controlled ship's state at the next moment, the control inputs of the guide ship and the propulsion ship, and the controlled ship's state at the current moment, the control inputs of the guide ship and the propulsion ship includes: The control input error of the guide ship and the propulsion ship is determined based on the control input of the guide ship and the propulsion ship at the next moment and the control input of the guide ship and the propulsion ship at the current moment. The state error of the controlled vessel is determined based on the state of the controlled vessel at the next moment and the state of the controlled vessel at the current moment, and the state errors of the guide vessel and the propulsion vessel are determined based on the state error of the controlled vessel. The power consumption of the guide vessel and the propulsion vessel is determined, and a cost function for nonlinear predictive model control is constructed based on the control input error, state error, and power consumption of the guide vessel and the propulsion vessel.
[0009] In one possible implementation, the cost function is: , in, The value of the cost function, For the state error associated with the guide vessel, For the state error associated with the propulsion vessel, Weights for the state errors associated with the guide vessel. As the weight of the state error associated with the propulsion vessel, To guide the control input error of the ship, To propel the ship's control input error, The weights of the control input errors for guiding the ship. The weights of the control input errors for propulsion vessels, The weighting of the power consumption of the guiding ship. Weighting of the power consumption of the propulsion ship. To guide the ship's power consumption, To reduce the power consumption of the propulsion ship, For transpose, For a moment.
[0010] In one possible implementation, the step of adaptively adjusting the weight parameters of the cost function to obtain the optimal solution of the cost function, with the goal of minimizing its value, includes: The weight parameters of the cost function are input into the constructed LightGBM model to obtain the optimal solution of the cost function.
[0011] In one possible implementation, the step of inputting the weight parameters of the cost function into the constructed LightGBM model to obtain the optimal solution of the cost function includes: The SHAP model is used to interpret the prediction results of the LightGBM model.
[0012] Secondly, the present invention also provides a ship berthing maneuvering control device, comprising: The prediction module is used to construct a three-degree-of-freedom ship maneuvering model in which the controlled ship, the guide ship, and the propulsion ship cooperate. Based on the three-degree-of-freedom ship maneuvering model, the reference path for the ship's berthing, and the current state of the controlled ship, a nonlinear prediction model is used to predict the state of the controlled ship and the control inputs of the guide ship and the propulsion ship, so as to obtain the state of the controlled ship and the control inputs of the guide ship and the propulsion ship at the next moment. The optimization module is used to construct a cost function for nonlinear predictive model control based on the controlled ship's state at the next moment, the control inputs of the guide ship and the propeller ship, and the current state of the controlled ship, the guide ship, and the propeller ship. With the goal of minimizing the value of the cost function, the weight parameters of the cost function are adaptively adjusted to obtain the optimal solution of the cost function. Based on the optimal solution, the optimal state of the controlled ship and the optimal control inputs of the guide ship and the propeller ship at the next moment are obtained. The control module is used to control the controlled ship based on the optimal state of the controlled ship at the next moment and the optimal control inputs of the guide ship and the propeller ship. Iterative updates are performed using the control inputs of the guide ship and the propeller ship at the next moment and the state of the controlled ship as the new state of the controlled ship and the control inputs of the guide ship and the propeller ship at the current moment, to achieve ship berthing operations.
[0013] Thirdly, the present invention also provides a shore-based ship navigation and control device, comprising: a processor and a memory; The memory stores a computer-readable program that can be executed by the processor; When the processor executes the computer-readable program, it implements the steps in the ship berthing maneuvering control method described above.
[0014] Thirdly, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instruction, which, when executed by a processor, can implement the steps of the ship berthing maneuvering control method described in any one of the above-mentioned method items.
[0015] The beneficial effects of this invention are as follows: It constructs a three-degree-of-freedom ship maneuvering model in which the controlled ship, guide ship, and propulsion ship cooperate, providing a high-precision prediction model for nonlinear predictive model control, thus ensuring precise control of the controlled ship. Based on the three-degree-of-freedom ship maneuvering model, the reference path for berthing, and the current state of the controlled ship, a nonlinear predictive model is used to predict the state of the controlled ship and the control inputs of the guide ship and propulsion ship, obtaining the state of the controlled ship and the control inputs of the guide ship and propulsion ship at the next moment. Based on the state of the controlled ship at the next moment, the control inputs of the guide ship and propulsion ship, and the state of the controlled ship at the current moment, a nonlinear predictive model is constructed. The cost function of the test model control is optimized by adaptively adjusting the weight parameters of the cost function to obtain the optimal solution. Based on the optimal solution, the optimal state of the controlled vessel and the optimal control inputs of the guide and propulsion vessels are obtained in the next time step. The controlled vessel is then controlled based on the optimal state of the controlled vessel and the optimal control inputs of the guide and propulsion vessels in the next time step. By adaptively adjusting the weight parameters of the cost function, the convergence speed of the cost function is accelerated, the state error of the controlled vessel and the control error of the guide and propulsion vessels during the control process are reduced, the dynamic adaptability of multi-vehicle cooperation is effectively improved, the efficiency of collaborative control is increased, and the efficiency and safety of vessel berthing are improved. Attached Figure Description
[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0017] Figure 1 A flowchart illustrating an embodiment of the ship berthing maneuvering control method provided by the present invention; Figure 2 A schematic diagram of a three-degree-of-freedom ship maneuvering model for the ship berthing maneuvering control method provided by the present invention; Figure 3 A schematic diagram of an embodiment of the ship berthing maneuvering control device provided by the present invention; Figure 4 This is a schematic diagram of an embodiment of the ship shore-based navigation control equipment provided by the present invention. Detailed Implementation
[0018] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.
[0019] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0020] This invention discloses a method, apparatus, device, and medium for controlling ship berthing and maneuvering, which can be used in a computer. The method, apparatus, or computer-readable storage medium involved in this invention can be integrated with the aforementioned equipment or can be relatively independent.
[0021] One specific embodiment of the present invention discloses a ship berthing maneuvering control method, which can be executed by a computer, specifically by one or more processors of the computer. For example... Figure 1 As shown, the ship berthing maneuvering control method includes: S101. Construct a three-degree-of-freedom ship maneuvering model in which the controlled ship, the guide ship, and the propulsion ship cooperate. Based on the three-degree-of-freedom ship maneuvering model, the reference path for the ship's berthing, and the state of the controlled ship at the current moment, use a nonlinear predictive model to predict the state of the controlled ship and the control inputs of the guide ship and the propulsion ship, and obtain the state of the controlled ship and the control inputs of the guide ship and the propulsion ship at the next moment. It should be noted that constructing a three-degree-of-freedom ship maneuvering model provides a foundation for multi-agent collaborative operations and offers a more accurate predictive model for nonlinear predictive model control (NMPC), ensuring precise control of the controlled vessel and improving the safety of ship berthing.
[0022] S102. Based on the state of the controlled ship at the next moment, the control inputs of the guide ship and the propulsion ship, and the state of the controlled ship at the current moment, the control inputs of the guide ship and the propulsion ship, a cost function for nonlinear predictive model control is constructed. With the minimum value of the cost function as the objective, the weight parameters of the cost function are adaptively adjusted to obtain the optimal solution of the cost function. Based on the optimal solution, the optimal state of the controlled ship at the next moment and the optimal control inputs of the guide ship and the propulsion ship are obtained. It should be noted that by adaptively adjusting the weight parameters of the cost function using the LightGBM model, the convergence speed of the cost function is accelerated, and the efficiency of collaborative control is improved.
[0023] S103. Based on the optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propulsion vessel, control the controlled vessel, and use the optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propulsion vessel as the new state of the controlled vessel at the current moment and the control inputs of the guide vessel and the propulsion vessel, and perform iterative updates to realize the berthing operation of the vessel.
[0024] In some embodiments, in step S101, a three-degree-of-freedom ship maneuvering model is constructed, in which the controlled ship, the guide ship, and the propulsion ship cooperate. For a schematic diagram of the three-degree-of-freedom ship maneuvering model, please refer to [link / reference needed]. Figure 2 ,like Figure 2 As shown, the left side is the established ship motion model, in which a geodetic coordinate system is introduced. Ship fixed coordinate system The right side shows a three-degree-of-freedom ship maneuvering model. While the ship's surface motion is a six-degree-of-freedom rigid body motion, in ship motion simulation and control, pitch, heave, and roll are generally not considered; the focus is more on the ship's motion within the horizontal plane, namely pitch, sway, and bow roll. Therefore, a three-degree-of-freedom ship maneuvering model is used to describe ship motion, such as... Figure 2 As shown, two tugboats control the vessel via thrust control. One tugboat, located closer to the bow of the controlled vessel, acts as the guide vessel, while the other acts as the propulsion vessel. The controlled vessel is a non-powered floating platform. The forces exerted on the controlled vessel by the guide vessel and the propulsion vessel are respectively the thrust exerted by the guide vessel on the controlled vessel. and thrust direction The thrust exerted by the propulsion vessel on the controlled vessel and thrust direction By applying thrust to the controlled vessel through the guide vessel and propulsion vessel, the state of the controlled vessel is altered to achieve multi-agent coordinated berthing operations. This involves multi-agent coordinated control to achieve berthing of the controlled vessel, where the multi-agents are the guide vessel, propulsion vessel, and the controlled vessel. The state of the controlled vessel includes its longitudinal speed. lateral velocity Ship position angular velocity of heading and bow roll rate The three-degree-of-freedom ship maneuvering model, in which the controlled vessel, guide vessel, and propulsion vessel work together, is as follows: , in, For the quality of the controlled vessel, , This is a coefficient related to the ship's hydrodynamics in longitudinal motion. For lateral velocity, This is a coefficient related to the ship's hydrodynamics in lateral motion. For longitudinal velocity, The angular velocity of the bow. The sum of the external moments generated in the longitudinal motion by the forces exerted on the controlled vessel by the guiding vessel and the propulsion vessel. The sum of the external moments generated by the forces exerted on the controlled vessel by the guiding vessel and the propulsion vessel on its lateral motion. The sum of the external torques generated by the forces exerted on the controlled vessel by the guiding vessel and the propulsion vessel on the bow motion. Let be the derivative of the bow angular velocity with respect to time. Let be the derivative of the transverse velocity with respect to time. The derivative of the longitudinal velocity with respect to time. The moment of inertia of the controlled vessel. The coefficient related to the ship's hydrodynamics in the bow rolling motion; Based on a three-degree-of-freedom ship maneuvering model, a reference path for ship berthing, and the current state of the controlled ship, a nonlinear predictive model is used to predict the state of the controlled ship and the control inputs of the guide and propulsion vessels. This predicts the state of the controlled ship and the control inputs of the guide and propulsion vessels at the next moment. The NMPC model is used for coordinated control during the coordinated berthing operation. The NMPC model has predictive capabilities and its behavior is closer to human behavior in coordinated berthing operations. First, the reference path for ship berthing is determined. Using key points of ship berthing as the data basis, cubic spline interpolation is performed between the key points to simulate a real continuous smooth navigation trajectory. Based on the set initial values, the specific values of all trajectory points are calculated sequentially at certain time intervals, thus obtaining the reference path for ship berthing. After determining the reference path, NMPC is used for prediction based on the three-degree-of-freedom ship maneuvering model, the reference path, and the current state of the controlled ship. Based on the given nonlinear system dynamics model, i.e., the three-degree-of-freedom ship maneuvering model, the continuous-time form of NMPC is: , The discrete-time model is obtained by discretization (sampling time is T): , in, For state vectors, To control the input vector, To describe the system's dynamics using a nonlinear function, based on the allocation of roles for tugboat cooperation in actual berthing operations, the two tugboats are divided into a guide vessel and a propulsion vessel, and the state of the controlled vessel is... Each transmission process will be divided into and , Represents the position and speed controlled by the propulsion vessel. The heading angle and bow roll rate, representing the control of the guide ship, can be expressed as follows: , in, The control sequence (a set of vector thrusts) output by the guiding ship. The control sequence (a set of vector thrust) output by the propulsion vessel. The NMPC can predict the state of the controlled vessel at the next moment, as well as the control inputs for the guide and propulsion vessels.
[0025] In some embodiments, in step S102, a cost function for nonlinear predictive model control is constructed based on the state of the controlled ship at the next moment, the control inputs of the guide ship and the propulsion ship, and the state of the controlled ship at the current moment, the control inputs of the guide ship and the propulsion ship. The control input errors of the guide ship and the propulsion ship are determined based on the control inputs of the guide ship and the propulsion ship at the next moment and the control inputs of the guide ship and the propulsion ship at the current moment. The state error of the controlled ship is determined based on the state error of the controlled ship at the next moment and the state of the controlled ship at the current moment. The state errors of the guide ship and the propulsion ship are determined based on the state errors of the controlled ship. The power consumption of the guide ship and the propulsion ship is determined. A cost function for nonlinear predictive model control is constructed based on the control input errors, state errors, and power consumption of the guide ship and the propulsion ship. To better demonstrate the agent-based nature of collaborative control and improve the independence and autonomy of the guide ship agent and the propulsion ship agent, the cost function is optimized into the form of agent application. The cost function is as follows: , in, The value of the cost function, For the state error associated with the guide vessel, For the state error associated with the propulsion vessel, Weights for the state errors associated with the guide vessel. As the weight of the state error associated with the propulsion vessel, To guide the control input error of the ship, To propel the ship's control input error, The weights of the control input errors for guiding the ship. The weights of the control input errors for propulsion vessels, The weighting of the power consumption of the guiding ship. Weighting of the power consumption of the propulsion ship. To guide the ship's power consumption, To reduce the power consumption of the propulsion ship, For transpose, For a specific moment; By setting a cost function To measure performance in the prediction time domain, the goal is to minimize the cost function. The weight parameters of the cost function are adaptively adjusted to obtain the optimal solution. The weight parameters of the cost function are: , , , , , The weight parameters of the cost function are input into the constructed LightGBM model to obtain the optimal solution of the cost function, i.e., to obtain the minimum cost function value. By introducing LightGBM (LightGradient Boosting Machine), the learning ability of the agent is improved, and the logical relationship between "weight parameters and cost" is explored. The goal of the LightGBM model is to update the model parameters by minimizing the objective function in each iteration. The objective function is usually defined as the gradient descent of the loss function. Specifically, for a given dataset... The goal of the LightGBM model is to find the parameters that minimize the objective function. The loss function of the LightGBM model is: , in, The loss function is (e.g., mean squared error or cross-entropy). For model parameters, For the dataset, For predicting output; During the model training phase, the cleaned dataset is divided into training and testing sets in an 8:2 ratio to ensure the model's generalization ability. The LightGBM model is initialized by setting initial hyperparameters such as learning rate, number of trees, and maximum depth. The model is then trained using the training set data, with the weight parameters as the input. , , , , , The output is the value of the cost function. LightGBM optimizes the loss function by iteratively constructing a decision tree. In each iteration, the model adjusts the parameters to minimize the prediction error. After obtaining the optimal solution of the cost function, the optimal state of the controlled ship at the next moment and the optimal control inputs of the guide ship and the propulsion ship are obtained based on the optimal solution.
[0026] The SHAP model interprets the prediction results of the LightGBM model. It employs the SHAP (SHapley Additive ex Planations) model for interpretability. The SHAP model is an additive interpretability model derived from the Shapley value concept in game theory. First, the SHAP model treats each feature as an independent contributor. After a sample is input into the model, each feature contributes to the prediction result for that sample. This contribution value is the classic SHAP value for that feature. The SHAP formula is as follows: , in, It is the first The first sample One characteristic, This is the classic SHAP value of this feature. Is the sample added With removal The difference between the model prediction results It is a reaction The weights of importance in different combinations Indicates the difference between features The feature subset other than the original input sample; after calculating the corresponding classical SHAP value, the Deep SHAP formula for interpreting deep learning models is selected to obtain the true SHAP value of the interpretable model. This calculation requires generating a new set of perturbation samples centered on the originally selected input sample. These perturbation samples are generated by making small, local changes to the navigation condition features and environmental features of the original sample. This is usually achieved by randomly sampling the features of the original sample or adding noise. The Deep SHAP formula is: , in, This is the true SHAP value of this feature. It is the difference between the features of the perturbed sample and the features of the original sample. It is the number of samples in the perturbation sample set. , These are the weights corresponding to the perturbation samples, and their magnitudes are usually determined based on the distance to the perturbation samples and the sample distribution. The numerical values illustrate the strength of the impact on model predictions. The sign indicates the nature of the impact on the model's predictions. This indicates that the feature has a positive impact on the model's predictions, meaning it improves the model's predicted values; if This feature has a negative impact on model predictions, meaning it reduces the predicted values of the model.
[0027] The visualization and interpretation of the SHAP model can be broken down into local and global interpretations. Local interpretation explains the features that affect the prediction results of a single sample. It can intuitively help humans understand the mechanism and logic of a single control process in collaborative control by attempting to enumerate the influence nature and intensity of individual sample features. Global interpretation explains the features that affect the prediction results of the entire sample set. It can design an overall feature importance ranking chart based on SHAP values to help humans understand the overall process of collaborative control. In addition, global interpretation considers the additional combined feature effects and can perform SHAP feature interaction analysis. Based on the SHAP interaction index of two features, a dependency graph is generated to show the marginal effect of the two features on the model prediction results.
[0028] In some embodiments, in step S103, the controlled vessel is controlled based on the optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propulsion vessel. The optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propulsion vessel are used as the new current state of the controlled vessel and the control inputs of the guide vessel and the propulsion vessel for iterative updates to achieve berthing operations. After receiving the optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propulsion vessel output by the NMPC, vector thrust control commands for the guide vessel and the propulsion vessel are obtained. These control commands are divided into... and (Each contains a set of vector thrust values), with the guide ship and propulsion ship independently executing control commands. The guide ship aims to control the heading angle and bow roll rate of the controlled vessel. Instructions The propulsion vessel executes actions aimed at controlling the position and speed of the controlled vessel. Instructions The current state of the controlled vessel after receiving the control command. By taking the optimal state of the controlled ship in the next moment and the optimal control inputs of the guide ship and propeller ship as the new state of the controlled ship in the current moment and the control inputs of the guide ship and propeller ship, the control output is continuously and dynamically adjusted according to the current state and future predictions to obtain the optimal control of the guide ship and propeller ship, while satisfying the constraint cost function. The condition that the value approaches the minimum is used to achieve ship berthing operations.
[0029] In the collaborative control process of the guide vessel, propulsion vessel, and controlled vessel—a multi-agent collaborative control process—a shore-based ship navigation control system is established. This system enables remote control of the vessel, monitoring of its status data and external environmental data, and maneuvering of the vessel. The shore-based system includes the controlled vessel, a cloud platform, and a shore-based control center. The controlled vessel's onboard sensing equipment perceives its own motion status and external environmental data, transmitting this information to the shore-based control center via the cloud platform. The guide vessel and propulsion vessel, based on the input of external information, output control commands for the current berthing control task. Furthermore, the remote controller can also autonomously issue corresponding control commands based on information perceived by the control center, serving as an emergency measure. The controlled vessel's onboard terminal consists of sensing equipment, a steering gear, and main engine control for data perception and shipboard control. The shore-based control center consists of a server, a telegraph handle, a steering gear handle, multiple displays, and a differential GPS base station for data monitoring and remote driving decision-making. The cloud platform consists of a communication system and a cloud server for data processing and transmission.
[0030] In summary, the ship berthing maneuvering control method provided by this invention constructs a three-degree-of-freedom ship maneuvering model in which the controlled ship, the guide ship, and the propulsion ship cooperate. Based on the three-degree-of-freedom ship maneuvering model, the ship's berthing reference path, and the current state of the controlled ship, a nonlinear predictive model is used to predict the state of the controlled ship and the control inputs of the guide ship and the propulsion ship, thereby obtaining the state of the controlled ship and the control inputs of the guide ship and the propulsion ship at the next moment. Based on the state of the controlled ship at the next moment, the control inputs of the guide ship and the propulsion ship, and the current state of the controlled ship and the control inputs of the guide ship and the propulsion ship, a nonlinear predictive model is constructed for control. The cost function aims to minimize its value. The weight parameters of the cost function are adaptively adjusted to obtain its optimal solution. Based on this optimal solution, the optimal state of the controlled vessel and the optimal control inputs for the guide and propulsion vessels are obtained for the next time step. The controlled vessel is then controlled based on these optimal state and control inputs. These optimal state and control inputs are then used as the new current state and control inputs for the guided and propulsion vessels, and the process is iteratively updated to achieve efficient and safe berthing operations.
[0031] To better implement the ship berthing maneuvering control method in the embodiments of the present invention, based on the ship berthing maneuvering control method, correspondingly, as follows: Figure 3 As shown, this embodiment of the invention also provides a ship berthing maneuvering control device, the ship berthing maneuvering control device 300 including: The prediction module 301 is used to construct a three-degree-of-freedom ship maneuvering model in which the controlled ship, the guide ship, and the propulsion ship cooperate. Based on the three-degree-of-freedom ship maneuvering model, the reference path for the ship's berthing, and the state of the controlled ship at the current moment, a nonlinear prediction model is used to predict the state of the controlled ship and the control inputs of the guide ship and the propulsion ship, so as to obtain the state of the controlled ship and the control inputs of the guide ship and the propulsion ship at the next moment. The optimization module 302 is used to construct a cost function for nonlinear predictive model control based on the state of the controlled vessel at the next moment, the control inputs of the guide vessel and the propulsion vessel, and the state of the controlled vessel at the current moment, the control inputs of the guide vessel and the propulsion vessel. With the goal of minimizing the value of the cost function, the weight parameters of the cost function are adaptively adjusted to obtain the optimal solution of the cost function. Based on the optimal solution, the optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propulsion vessel are obtained. The control module 303 is used to control the controlled vessel based on the optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propulsion vessel. The control inputs of the guide vessel and the propulsion vessel at the next moment and the state of the controlled vessel are used as the new state of the controlled vessel at the current moment and the control inputs of the guide vessel and the propulsion vessel for iterative updates to achieve the vessel berthing operation.
[0032] like Figure 4 As shown, the present invention also provides a shore-based ship navigation control device 400, which can be a mobile terminal, desktop computer, laptop, handheld computer, server, or other computing device. The shore-based ship navigation control device 400 includes a processor 401, a memory 402, and a display 403. Figure 4 Some components of the shore-based navigation system 400 are shown, but it should be understood that implementation of all shown components is not required, and more or fewer components may be implemented instead.
[0033] In some embodiments, the memory 402 may be an internal storage unit of the ship-to-shore navigation control equipment 400, such as a hard disk or memory of the ship-to-shore navigation control equipment 400. In other embodiments, the memory 402 may be an external storage device of the ship-to-shore navigation control equipment 400, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the ship-to-shore navigation control equipment 400. Further, the memory 402 may include both internal and external storage units of the ship-to-shore navigation control equipment 400. The memory 402 is used to store application software and various types of data installed on the ship-to-shore navigation control equipment 400, such as the program code installed on the ship-to-shore navigation control equipment 400. The memory 402 may also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 402 stores a ship berthing maneuvering control program, which can be executed by the processor 401 to implement the ship berthing maneuvering control method of the various embodiments of the present invention.
[0034] In some embodiments, processor 401 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 402 or process data, such as ship berthing maneuvering control methods.
[0035] In some embodiments, display 403 may be an LED display, a liquid crystal display, a touch-screen liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 403 is used to display identification information of the ship's berthing maneuvering control program and to display a visual user interface. Components 401-403 of the ship's shore-based navigation equipment 400 communicate with each other via a system bus.
[0036] In some embodiments, when the processor 401 executes the ship berthing maneuvering control program in the memory 402, it implements the various steps of the ship berthing maneuvering control method as described in the above embodiments. Since the ship berthing maneuvering control method has been described in detail above, it will not be repeated here.
[0037] Accordingly, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instruction, which, when executed by a processor, can realize the steps or functions of the ship berthing maneuvering control method provided in the above-described method embodiments.
[0038] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.
[0039] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for controlling the berthing and maneuvering of a ship, characterized in that, include: A three-degree-of-freedom ship maneuvering model is constructed, in which the controlled ship, the guide ship, and the propulsion ship cooperate. Based on the three-degree-of-freedom ship maneuvering model, the reference path for the ship's berthing, and the current state of the controlled ship, a nonlinear predictive model is used to predict the state of the controlled ship and the control inputs of the guide ship and the propulsion ship, so as to obtain the state of the controlled ship and the control inputs of the guide ship and the propulsion ship at the next moment. Based on the state of the controlled ship at the next moment, the control inputs of the guide ship and the propulsion ship, and the state of the controlled ship at the current moment, the cost function of the nonlinear predictive model control is constructed. With the goal of minimizing the value of the cost function, the weight parameters of the cost function are adaptively adjusted to obtain the optimal solution of the cost function. Based on the optimal solution, the optimal state of the controlled ship at the next moment and the optimal control inputs of the guide ship and the propulsion ship are obtained. The controlled vessel is controlled based on the optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propulsion vessel. The optimal state of the controlled vessel at the next moment and the optimal control inputs of the guide vessel and the propulsion vessel are used as the new state of the controlled vessel at the current moment and the control inputs of the guide vessel and the propulsion vessel, and the process is iteratively updated to achieve the vessel berthing operation.
2. The ship berthing maneuvering control method according to claim 1, characterized in that, The three-degree-of-freedom ship maneuvering model is as follows: ， in, For the quality of the controlled vessel, , This is a coefficient related to the ship's hydrodynamics in longitudinal motion. For lateral velocity, This is a coefficient related to the ship's hydrodynamics in lateral motion. For longitudinal velocity, The angular velocity of the bow. The sum of the external moments generated in the longitudinal motion by the forces exerted on the controlled vessel by the guiding vessel and the propulsion vessel. The sum of the external moments generated by the forces exerted on the controlled vessel by the guiding vessel and the propulsion vessel on its lateral motion. The sum of the external torques generated by the forces exerted on the controlled vessel by the guiding vessel and the propulsion vessel on the bow motion. Let be the derivative of the bow angular velocity with respect to time. Let be the derivative of the transverse velocity with respect to time. The derivative of the longitudinal velocity with respect to time. The moment of inertia of the controlled vessel. This is a coefficient related to the ship's hydrodynamics during bow rolling motion.
3. The ship berthing maneuvering control method according to claim 2, characterized in that, The state of the controlled vessel includes longitudinal speed, lateral speed, vessel position, heading angular velocity, and bow roll angular velocity. The control inputs include the thrust and thrust direction exerted by the guide vessel on the controlled vessel and the thrust and thrust direction exerted by the propulsion vessel on the controlled vessel.
4. The ship berthing maneuvering control method according to claim 3, characterized in that, The cost function for constructing the nonlinear predictive model control based on the controlled ship's state at the next moment, the control inputs of the guide ship and the propulsion ship, and the current moment's controlled ship state, the guide ship and the propulsion ship includes: The control input error of the guide ship and the propulsion ship is determined based on the control input of the guide ship and the propulsion ship at the next moment and the control input of the guide ship and the propulsion ship at the current moment. The state error of the controlled vessel is determined based on the state of the controlled vessel at the next moment and the state of the controlled vessel at the current moment, and the state errors of the guide vessel and the propulsion vessel are determined based on the state error of the controlled vessel. The power consumption of the guide vessel and the propulsion vessel is determined, and a cost function for nonlinear predictive model control is constructed based on the control input error, state error, and power consumption of the guide vessel and the propulsion vessel.
5. The ship berthing maneuvering control method according to claim 4, characterized in that, The cost function is: ， in, The value of the cost function, For the state error associated with the guide vessel, For the state error associated with the propulsion vessel, Weights for the state errors associated with the guide vessel. As the weight of the state error associated with the propulsion vessel, To guide the control input error of the ship, To propel the ship's control input error, The weights of the control input errors for guiding the ship. The weights of the control input errors for propulsion vessels, The weighting of the power consumption of the guiding ship. Weighting of the power consumption of the propulsion ship. To guide the ship's power consumption, To reduce the power consumption of the propulsion ship, For transpose, For a moment.
6. The ship berthing maneuvering control method according to claim 4, characterized in that, The step of adaptively adjusting the weight parameters of the cost function to obtain the optimal solution for the cost function, with the goal of minimizing its value, includes: The weight parameters of the cost function are input into the constructed LightGBM model to obtain the optimal solution of the cost function.
7. The ship berthing maneuvering control method according to claim 6, characterized in that, The step of inputting the weight parameters of the cost function into the constructed LightGBM model to obtain the optimal solution of the cost function includes: The SHAP model is used to interpret the prediction results of the LightGBM model.
8. A ship berthing maneuvering control device, characterized in that, include: The prediction module is used to construct a three-degree-of-freedom ship maneuvering model in which the controlled ship, the guide ship, and the propulsion ship cooperate. Based on the three-degree-of-freedom ship maneuvering model, the reference path for the ship's berthing, and the current state of the controlled ship, a nonlinear prediction model is used to predict the state of the controlled ship and the control inputs of the guide ship and the propulsion ship, so as to obtain the state of the controlled ship and the control inputs of the guide ship and the propulsion ship at the next moment. The optimization module is used to construct a cost function for nonlinear predictive model control based on the controlled ship's state at the next moment, the control inputs of the guide ship and the propeller ship, and the current state of the controlled ship, the guide ship, and the propeller ship. With the goal of minimizing the value of the cost function, the weight parameters of the cost function are adaptively adjusted to obtain the optimal solution of the cost function. Based on the optimal solution, the optimal state of the controlled ship and the optimal control inputs of the guide ship and the propeller ship at the next moment are obtained. The control module is used to control the controlled ship based on the optimal state of the controlled ship at the next moment and the optimal control inputs of the guide ship and the propeller ship. Iterative updates are performed using the control inputs of the guide ship and the propeller ship at the next moment and the state of the controlled ship as the new state of the controlled ship and the control inputs of the guide ship and the propeller ship at the current moment, to achieve ship berthing operations.
9. A shore-based ship navigation and control device, characterized in that, Including memory and processor; The memory stores a computer-readable program that can be executed by the processor; When the processor executes the computer-readable program, it implements the steps in the ship berthing maneuvering control method as described in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the ship berthing maneuvering control method according to any one of claims 1-7.