Ship hybrid formation control method, device, equipment and medium

Abstract

The invention relates to a ship hybrid formation control method, device and equipment and a medium, and belongs to the technical field of ships, and the ship hybrid formation control method comprises the steps: constructing a hybrid formation of unmanned ships guided by a man-driven ship through a LoRa technology, obtaining the dynamic data of the hybrid formation, fusing the dynamic data, and obtaining the hybrid formation based on the fused dynamic data. An EMD-LSTM network is adopted to predict the navigation trajectory of the mixed formation to obtain a predicted trajectory, a navigation risk early warning system is constructed based on the obtained actual trajectory of the mixed formation and the predicted trajectory, the navigation risk of the mixed formation is evaluated based on the navigation risk early warning system to obtain an evaluation result, and the mixed formation is controlled based on the evaluation result. And the formation decision-making efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the field of ship technology, and in particular to a method, device, equipment and medium for controlling a mixed ship formation. Background Art

[0002] During the transformation period of the shipping industry, the coordinated formation of unmanned and manned ships has become a hot topic of exploration. With the continuous improvement of autonomous navigation technology, there will be more and more autonomous ships in the shipping field in the future, and their role in reducing labor costs, improving transportation flexibility and promoting industry intelligence will become more and more significant. However, despite the many advantages of autonomous ships, single unmanned ships still face challenges in autonomous decision-making in the face of complex and changing navigation environments. In order to overcome this problem, it is crucial to combine the flexible response advantages of manned ships with the efficient operation capabilities of unmanned ships to build a hybrid formation strategy with complementary advantages. This strategy aims to comprehensively improve the overall efficiency of shipping by integrating the advantages of both parties to achieve more efficient and safer navigation. Looking to the future, with the continuous advancement of technology and the increasing richness of application scenarios, the mixed formation mode of manned and unmanned ships will become the mainstream of the shipping industry. This model will lead the shipping industry to a higher level of intelligence and efficiency, and provide more reliable and efficient transportation services for global trade.

[0003] However, when establishing a mixed formation transport system, manned ships usually use traditional radio communications, while unmanned ships prefer satellite communications. This difference leads to communication frequency mismatch and technical incompatibility problems, increasing the uncertainty of formation decision-making; and when the mixed formation conducts scientific quantification and evaluation of possible risks during navigation, there is a lack of effective navigation risk assessment standards, resulting in low efficiency and accuracy of formation decision-making, which affects the safety and efficiency of shipping operations. Summary of the invention

[0004] In view of this, it is necessary to provide a method, device, equipment and medium for controlling a mixed formation of ships, so as to solve the technical problem that a mixed formation of manned ships leading unmanned ships lacks effective navigation risk assessment standards during shipping operations, resulting in low efficiency and accuracy of formation decision-making.

[0005] In order to solve the above problems, the present invention provides a mixed formation control method for ships, comprising: Use LoRa technology to build a mixed fleet of manned ships leading unmanned ships; acquiring dynamic data of the mixed formation, fusing the dynamic data, and predicting the navigation trajectory of the mixed formation using an EMD-LSTM network based on the fused dynamic data to obtain a predicted trajectory; A navigation risk warning system is constructed based on the actual trajectory of the mixed formation and the predicted trajectory, the navigation risk of the mixed formation is evaluated based on the navigation risk warning system, an evaluation result is obtained, and the mixed formation is controlled based on the evaluation result.

[0006] In a possible implementation, the construction of a mixed formation of a manned ship leading an unmanned ship by using LoRa technology includes: With manned ships as the LoRa hub and unmanned ships as the LoRa terminals, a mixed fleet of manned ships leading unmanned ships is built.

[0007] In a possible implementation manner, fusing the dynamic data includes: Acquiring dynamic data of the mixed formation during navigation through AIS, radar, GNSS and electronic compass, wherein the dynamic data at least includes ship position, heading, navigation speed, AIS track and radar track; After refining the dynamic data, refined dynamic data is obtained; The MMSI of the ship is obtained. Based on the refined data, the ship position, heading and AIS are associated through the MMSI of the ship, the AIS is optimized, and the optimized AIS tracks are clustered using DBSCAN. The clustered AIS tracks and radar tracks are bidirectionally aligned using the BDTW algorithm to complete the fusion of dynamic data.

[0008] In a possible implementation, the EMD-LSTM network includes an EMD algorithm and an LSTM network; the method of using the EMD-LSTM network to predict the navigation trajectory of the mixed formation based on the fused dynamic data to obtain the predicted trajectory includes: Obtaining historical trajectory characteristics of the ships in the mixed formation based on the fused dynamic data, wherein the historical trajectory characteristics at least include the longitude, latitude, speed and heading of the ships; Inputting the historical trajectory features into the LSTM network, predicting the navigation trajectory of the mixed formation through the LSTM network, and obtaining a first predicted trajectory and a second predicted trajectory; Acquire an actual trajectory of the mixed formation, determine a residual sequence based on the actual trajectory and the first predicted trajectory, decompose the residual sequence by an EMD algorithm, and obtain IMFs and residual terms; Inputting the IMFs and residual items into the LSTM network, predicting the IMFs and residual items through the LSTM network, and obtaining prediction results of the IMFs and residual items; The prediction results of the IMFs and the prediction results of the residual items are superimposed respectively, and a prediction trajectory is obtained based on the second prediction trajectory, the superimposed IMFs and the superimposed residual items.

[0009] In a possible implementation, the navigation risk warning system includes a collision risk index, a deviation risk index, and a grounding risk index; and constructing the navigation risk warning system based on the actual trajectory of the mixed formation and the predicted trajectory obtained includes: Preset a minimum safety interval threshold, determine the positions of the ships in the mixed formation in the future time based on the predicted trajectory, calculate the minimum interval distance between the ships based on the positions of the ships, and determine a first collision risk indicator based on the minimum safety interval threshold and the minimum interval distance, wherein the future time is three minutes; Determining a safety zone for the ships in the mixed formation, determining a location of an obstacle based on the actual trajectory and the predicted trajectory, and determining a second collision risk indicator based on the safety zone for the ships in the mixed formation and the location of the obstacle; determining a collision risk indicator based on the first collision risk indicator and the second collision risk indicator; Preset a safety distance threshold, determine the maximum distance between the unmanned ship and the manned ship based on the actual trajectory and the predicted trajectory, and determine a collision risk index based on the maximum distance between the unmanned ship and the manned ship and the safety distance threshold; The electronic chart information is obtained, and a ship draft margin is preset. The actual draft and maximum draft of the ships in the mixed formation are determined based on the actual trajectory, the predicted trajectory and the electronic chart information. The grounding risk index is determined based on the ship draft margin, the maximum draft and the ship draft margin.

[0010] In a possible implementation, the calculation formula of the collision risk index is: , , , in, is the collision risk indicator, is the first collision risk indicator, is the second collision risk indicator, The minimum distance between the predicted positions of ships in the fleet within three minutes. is the minimum safety interval threshold, is the location of the obstacle, For the mixed formation The scope of the safety area of ​​the ship; The calculation formula of the deviation risk index is: , in, is the deviation risk indicator, For the The maximum distance between an unmanned ship and the pilot ship, is the preset safety distance threshold; The calculation formula of the stranding risk index is: , in, is the stranding risk indicator, For the mixed formation The actual draft of the ship, is the preset draft margin, The maximum draft allowed for the navigation path.

[0011] In a possible implementation manner, controlling the mixed formation based on the evaluation result includes: The navigation status of the mixed formation is judged based on the evaluation result. When the navigation status of the mixed formation is abnormal, alarm information is obtained. Based on the alarm information, adjustment instructions are issued through the LoRa communication network. Based on the adjustment instructions, the navigation parameters of the unmanned ship are adjusted to complete the control of the mixed formation.

[0012] On the other hand, the present invention also provides a mixed formation control device for ships, comprising: A mixed formation building module, which is used to build a mixed formation of manned ships leading unmanned ships through LoRa technology; A trajectory prediction module is used to obtain dynamic data of the mixed formation, fuse the dynamic data, and then predict the navigation trajectory of the mixed formation using an EMD-LSTM network based on the fused dynamic data to obtain a predicted trajectory; A mixed formation control module is used to build a navigation risk warning system based on the actual trajectory of the mixed formation and the predicted trajectory, evaluate the navigation risk of the mixed formation based on the navigation risk warning system, obtain an evaluation result, and control the mixed formation based on the evaluation result.

[0013] In another aspect, the present invention further provides an electronic device, comprising: a processor and a memory; The memory stores a computer-readable program executable by the processor; When the processor executes the computer-readable program, the steps in the method for controlling a mixed fleet of ships as described above are implemented.

[0014] On the other hand, the present invention also provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the ship mixed formation control method as described above.

[0015] The beneficial effects of the present invention are as follows: a mixed formation of manned ships leading unmanned ships is constructed through the LoRa technology, and the communication between manned ships and unmanned ships is integrated by using the LoRa technology, so as to solve the problems of communication frequency mismatch and technical incompatibility, improve communication efficiency and compatibility, and ensure accurate transmission of information between manned ships and unmanned ships; dynamic data of the mixed formation is obtained, and the dynamic data is fused, so as to provide a more comprehensive and accurate decision-making basis for the mixed formation through data fusion; based on the fused dynamic data, the EMD-LSTM network is used to predict the navigation trajectory of the mixed formation, and by introducing the EMD-LSTM network, a high-precision prediction of the ship trajectory is achieved; a navigation risk warning system is constructed based on the actual trajectory and predicted trajectory of the obtained mixed formation, and the navigation risk of the mixed formation is evaluated based on the navigation risk warning system, and the risks occurring in the mixed formation during navigation are scientifically quantified and evaluated through the navigation risk warning system, dynamically responding to the instant changes in ship behavior, more sensitively identifying abnormally sailing ships, providing timely and accurate decision support for drivers, simplifying navigation complexity, and improving the efficiency and accuracy of formation decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A flow chart of an embodiment of a mixed fleet control method provided by the present invention; Figure 2 A schematic diagram of a ship formation communication network for the ship mixed formation control method provided by the present invention; Figure 3 A schematic structural diagram of an embodiment of a mixed formation control device for ships provided by the present invention; Figure 4 A schematic structural diagram of an embodiment of an electronic device provided by the present invention. DETAILED DESCRIPTION

[0017] The preferred embodiments of the present invention are described in detail below in conjunction with the accompanying drawings, wherein the accompanying drawings constitute a 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 used to limit the scope of the present invention.

[0018] The present invention discloses a method, device, equipment and medium for controlling a mixed fleet of ships, which can be used in a computer. The method, device or computer-readable storage medium involved in the present invention can be integrated with the above-mentioned device or can be relatively independent.

[0019] A specific embodiment of the present invention discloses a method for controlling a mixed ship formation, which can be executed by a computer, and specifically by one or more processors of the computer. Figure 1 As shown, the ship mixed formation control method includes: S101, using LoRa technology to build a mixed formation of manned ships leading unmanned ships; S102, acquiring dynamic data of the mixed formation, fusing the dynamic data, and predicting the navigation trajectory of the mixed formation using an EMD-LSTM network based on the fused dynamic data to obtain a predicted trajectory; S103: construct a navigation risk warning system based on the actual trajectory and predicted trajectory of the mixed formation, evaluate the navigation risk of the mixed formation based on the navigation risk warning system, obtain an evaluation result, and control the mixed formation based on the evaluation result.

[0020] Among them, the ship formation communication network is built through LoRa technology. The ship formation communication network is composed of LoRa hub, data processing center and terminal, forming a comprehensive communication network covering mixed formations and realizing seamless integration of data collection and navigation. Among them, manned ships are LoRa hubs and unmanned ships are LoRa terminals.

[0021] Compared with the prior art, the mixed formation control method of ships provided in this embodiment uses LoRa technology to build a mixed formation in which manned ships lead unmanned ships, integrates the communication between manned ships and unmanned ships through LoRa technology, solves the problems of communication frequency mismatch and technical incompatibility, improves communication efficiency and compatibility, and ensures accurate transmission of information between manned ships and unmanned ships; obtains dynamic data of the mixed formation, fuses the dynamic data, and provides a more comprehensive and accurate decision-making basis for the mixed formation through data fusion; based on the fused dynamic data, an EMD-LSTM network is used to predict the navigation trajectory of the mixed formation to obtain a predicted trajectory, thereby achieving high-precision prediction of the ship trajectory; a navigation risk warning system is constructed based on the actual trajectory and predicted trajectory of the obtained mixed formation; based on the navigation risk warning system, the navigation risk of the mixed formation is evaluated to obtain an evaluation result; the mixed formation is controlled based on the evaluation result; the risks occurring in the mixed formation during navigation are scientifically quantified and evaluated through the navigation risk warning system, abnormally sailing ships are identified, and timely and accurate decision support is provided to the driver, thereby simplifying the navigation complexity and improving the efficiency and accuracy of formation decision-making.

[0022] In some embodiments, in step S101, a mixed formation of manned ships leading unmanned ships is constructed by LoRa technology, and a mixed formation of manned ships leading unmanned ships is constructed with manned ships as LoRa hubs and unmanned ships as LoRa terminals. In the mixed formation of manned ships leading unmanned ships, a ship formation communication network is constructed by LoRa technology, and manned ships lead the navigation of unmanned ships through the ship formation communication network. The ship formation communication network is composed of a LoRa hub, a data processing center and a terminal to form a comprehensive communication network covering the mixed formation. For a schematic diagram of the ship formation communication network, please refer to Figure 2 ,like Figure 2 As shown, the pilot ship (man-driven ship) is equipped with a LoRa server and a LoRa gateway, which is the LoRa hub. Each unmanned ship is equipped with a LoRa terminal. The LoRa hub equipped on the pilot ship is the information exchange core, responsible for data collection, transfer and management, to ensure the smooth flow of data from each ship. At the same time, the shipboard data processing center (LoRa server) is responsible for the in-depth analysis, intelligent evaluation and visual presentation of the data, providing the navigator with real-time formation status and optimizing the decision-making process. The unmanned ship is equipped with a LoRa terminal to automatically collect and upload key navigation data, and receive execution instructions at the same time to ensure coordination and flexible adjustment of the formation. The formation ships transmit position coordinates, heading angles and identification codes (MMSI) to the manned ships in real time through the LoRa terminals. The information is gathered on the server to ensure the dynamic and instantaneous sharing of the formation. By equipping the manned ships with LoRa hubs and the unmanned ships with LoRa terminals, a communication system covering the entire formation is collaboratively built to achieve seamless integration of data collection and navigation, and enhance the coordination and efficiency of teamwork.

[0023] In some embodiments, in step S102, dynamic data of the mixed formation is obtained, and dynamic data of the mixed formation during navigation is obtained through AIS, radar, GNSS and electronic compass, wherein the dynamic data at least includes ship position, heading, navigation speed, AIS track, radar track, and on a manned ship, the AIS system, radar monitoring, global navigation satellite system (GNSS) and electronic compass are comprehensively used to fully capture navigation dynamic data. The dynamic data is multi-dimensional data and is also the core data of the mixed formation navigation; the dynamic data is fused, and in the mixed formation navigation During the process, each ship dynamically generates core data such as positioning, heading, and navigation speed. These data are of core value to collaborative efficiency and decision-making. In order to achieve efficient integration of multi-source information, it is necessary to process heterogeneous data to ensure the unity and coherence of data in time, space, and meaning. This fusion process mainly involves data refinement, information linking, track improvement and intelligent analysis, track synchronization and optimization, and chart mapping and visualization to achieve accurate data integration and utilization. The specific steps are: First, refine the dynamic data to obtain the refined dynamic data, and remove the dynamic data of each ship in the formation through data refinement. Redundancy and outliers in ship data are streamlined and corrected to ensure the purity and consistency of the data set. The MMSI of the ship is obtained. Based on the refined data, the ship's position, heading and AIS are associated through the ship's MMSI. The real-time position, heading and AIS signal are associated in the LoRa information center with the ship's MMSI as the link to ensure the continuity and accuracy of the information flow. Secondly, the AIS track is optimized, and the AIS track is optimized using high-frequency positioning and heading data. The track gaps are filled through intelligent algorithms to enhance the track integrity and accuracy, and DBSCAN is used. The optimized AIS tracks are clustered, and DBSCAN is used for iterative clustering to simplify and highlight feature points, intelligently identify navigation modes, and improve data analysis and monitoring efficiency. The BDTW algorithm is used to perform bidirectional alignment of clustered AIS tracks and radar tracks, taking into account speed fluctuations, calculating the optimal matching path, and measuring similarity to complete the fusion of dynamic data and achieve accurate fusion of multi-source information. Finally, the electronic chart is extracted from the LoRa information center, and the ship dynamics after matching verification are intuitively presented on the chart, expanding the information perspective and ensuring the visualization and efficiency of data fusion. DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a typical density-based spatial clustering algorithm that discovers clusters of any shape and size through the concept of density. The steps of DBSCAN iterative clustering are: First, data preprocessing: collect ship trajectory data from the AIS system, including timestamp, position (longitude, latitude), speed, heading and other information, clean the data, remove outliers, duplicate values ​​and missing values, ensure the accuracy and completeness of the data, synchronize the AIS time, extract key features such as position, speed, heading, etc. from the preprocessed data, these features will be used for subsequent calculations and matching, secondly, initialize the AIS trajectory and initialize the core object set It is an empty set, the number of clusters k is 0, and the sample set is not visited The cluster partition C is an empty set, and the dataset D corresponds to the AIS trajectory data. For each sample in the dataset D, the distance metric is used to find the sample’s Neighborhood subsample set , when the number of samples in the subsample set satisfies , the sample is added to the core object sample set In the core object collection In the process, randomly select a core object o and initialize the core object queue of the current cluster , Category No. , the current cluster sample set , update the unvisited sample set , if the current cluster core object queue Not empty, take out a core object ,turn up All Neighborhood points, if a point p among these neighborhood points has not been classified, it will be added to the current cluster sample set If point p is a core object, add it Update the unvisited sample set , and repeat this process until If it is empty, the final cluster partition C is output, and the cluster partition C contains the trajectory feature points of AIS; The BDTW algorithm is an innovative time series analysis method, which is particularly suitable for matching trajectory data with different speeds and lengths. BDTW introduces the concept of bidirectional alignment, which allows two time series to be aligned not only in one direction (from the beginning to the end), but also in the opposite direction. This bidirectional alignment mechanism enables BDTW to more comprehensively capture the relative speed changes and morphological differences between the two sequences, thereby providing more accurate and robust trajectory matching performance. In ship trajectory matching, the BDTW algorithm can efficiently fuse AIS (Automatic Identification System) and radar trajectory data to provide more comprehensive and accurate data support for ship monitoring systems. The fusion steps are as follows: First, data preprocessing, collecting ship trajectory data from AIS and radar systems, including timestamps, locations (longitude, latitude), speed, heading and other information, cleaning the data, removing outliers, duplicate values ​​and missing values ​​to ensure the accuracy and completeness of the data, and at the same time, performing time synchronization processing on AIS and radar data to ensure that the two are comparable in the time dimension; Feature extraction: extract key features such as position, speed, heading, etc. from the preprocessed data. These features will be used for subsequent calculations and matching. Secondly, use the BDTW algorithm to calculate the similarity between the AIS and radar tracks. Through dynamic programming, BDTW finds the optimal path between the two time series so that they are as similar as possible in morphology. The bidirectional alignment mechanism of BDTW can fully capture the relative speed changes and morphological differences between the two sequences, thereby providing more accurate and robust track matching results. Finally, result optimization: perform a preliminary match between the AIS and radar tracks based on the calculation results, and use the optimization algorithm to further adjust the matching parameters to optimize the matching results. The accuracy and reliability of the matching results are verified through visualization and statistical test methods.

[0024] Based on the fused dynamic data, the EMD-LSTM network is used to predict the navigation trajectory of the mixed formation to obtain the predicted trajectory. In the navigation environment of the mixed formation, facing potential threats such as collision, course deviation and reefing, the EMD-LSTM network model is used to comprehensively analyze the key data such as the ship's position, heading, speed and heading, predict the future trajectory of the ship, and project the prediction results onto the electronic nautical chart in real time to provide decision support for the crew; the EMD-LSTM network includes the EMD algorithm and the LSTM network. The process of using the EMD-LSTM network to predict the navigation trajectory of the mixed formation is as follows: first, based on the fused dynamic data, the historical trajectory characteristics of the ships in the mixed formation are obtained, and the historical trajectory characteristics at least include the longitude, latitude, speed and heading of the ship, and the historical trajectory characteristics are input into the LSTM network. Trajectory characteristics of time It is expressed as: , in, For ships in The latitude of the moment, For ships in The longitude of the moment, For ships in The speed of the moment, For ships in The direction of the moment; will continue Ship navigation trajectory characteristic data at each moment and As the input of the EMD-LSTM network, Flight track data at the moment As output, Corresponding to the input layer Size, the expression of EMD-LSTM network is: , The navigation trajectory of the mixed formation is predicted through the LSTM network, and the training set and test set are used for prediction respectively to obtain the first predicted trajectory And the second prediction trajectory , the first predicted trajectory is the training set prediction value, the second predicted trajectory is the test set prediction value, and then the actual trajectory of the mixed formation is obtained. , based on the actual trajectory and the first predicted trajectory Determine the residual sequence , its residual sequence is the difference between the first predicted trajectory and the actual trajectory, and its calculation formula is: , The residual sequence is analyzed by EMD algorithm. Decompose and obtain IMFs and residual terms. Its IMFs are Eigenmode components , the residual term is , EMD decomposition can better capture the scale characteristics of the error at different times; finally, the IMFs and residual terms are input into the LSTM network, and the IMFs and residual terms are predicted by the LSTM network to obtain the prediction results of the IMFs and residual terms, for the intrinsic mode components of different frequencies and the residual The LSTM networks are established for training and prediction respectively to obtain the prediction results. The prediction results are used as the intrinsic mode components and residual terms of the test set prediction value and the actual trajectory residual. The prediction results of IMFs and the prediction results of the residual terms are superimposed respectively, that is, the prediction results are summed to obtain the prediction value of the test set residual. , based on the first prediction trajectory, the superimposed IMFs and the superimposed residual term, the prediction value of the test set residual is obtained. And the predicted value of the test set are combined to obtain the predicted trajectory, and the calculation formula of the predicted trajectory is: , in, To predict the trajectory; In the compilation phase of the EMD-LSTM network, the loss function and optimizer are defined to measure the difference between the model prediction results and the actual results and to minimize the defined loss function using an appropriate optimization algorithm. The mean square error (MSE) is selected as the loss function, and the loss function is calculated as follows: , in, is the loss function, is the real trajectory, To predict the trajectory, is the sample size.

[0025] In some embodiments, in step S103, a navigation risk warning system is constructed based on the actual trajectory and predicted trajectory of the obtained mixed formation, and the navigation risk warning system includes a collision risk index, a deviation risk index and a grounding risk index, and the process is: presetting a minimum safety interval threshold, determining the position of the ships in the mixed formation in the future time based on the predicted trajectory, calculating the minimum interval distance between the ships based on the position of the ships, and determining the first collision risk index based on the minimum safety interval threshold and the minimum interval distance, wherein the future time is three minutes, and the calculation formula of the first collision risk index is: , in, is the first collision risk index, i.e., the collision risk index within the mixed formation. The minimum distance between the predicted positions of ships in the fleet within three minutes. is the minimum safety interval threshold; Determine the safety area of ​​the ships in the mixed formation, determine the location of the obstacle based on the actual trajectory and the predicted trajectory, and determine the second collision risk index based on the safety area of ​​the ships in the mixed formation and the location of the obstacle. The calculation formula of the second collision risk index is: , in, is the second collision risk index, i.e., the collision risk index between the mixed formation and external obstacles. is the location of the obstacle, For the mixed formation The scope of the safety area of ​​the ship; The collision risk index is determined based on the first collision risk index and the second collision risk index. The calculation formula of the collision risk index is: , in, is a collision risk indicator; A safety distance threshold is preset, and the maximum distance between the unmanned ship and the manned ship is determined based on the actual trajectory and the predicted trajectory. The collision risk index is determined based on the maximum distance between the unmanned ship and the manned ship and the safety distance threshold. The calculation formula of the collision risk index is: , in, is the deviation risk indicator, For the The maximum distance between an unmanned ship and the pilot ship, The preset safety distance threshold Obtain electronic chart information, preset the ship draft margin, determine the actual draft and maximum draft of the ships in the mixed formation based on the actual track, predicted track and electronic chart information, and determine the grounding risk index based on the ship draft margin, maximum draft and ship draft margin. The calculation formula for the grounding risk index is: , in, For the risk of stranding, For the mixed formation The actual draft of the ship, is the preset draft margin, the maximum draught allowed for the navigation path; The navigation risk of the mixed formation is evaluated based on the navigation risk warning system to obtain the evaluation results, which are: there is a collision risk, there is no collision risk, there is a deviation risk, there is no deviation risk, there is a grounding risk and there is no grounding risk; during the navigation of the mixed formation, when the minimum interval distance of the predicted positions of each ship in the mixed formation in the next 3 minutes is less than the minimum safety interval threshold, that is, the collision risk index within the ship formation is less than 0, it means that there is a collision risk between the ships in the mixed formation, and when the minimum interval distance of the predicted positions of each ship in the mixed formation in the next 3 minutes is greater than or equal to the minimum safety interval threshold, there is no collision risk between the ships in the mixed formation; during navigation, when an obstacle is located within the safety area of ​​a ship in the mixed formation, it indicates that there is a collision risk, that is, When the obstacle is not within the safety area of ​​any ship, it means there is no collision risk. is 1; when the maximum distance between the unmanned ship and the manned ship is greater than the safety distance threshold, it indicates that there is a risk of deviation; when the maximum distance between the unmanned ship and the manned ship is less than or equal to the safety distance threshold, it indicates that there is no risk of deviation; during navigation, when the actual draft of the ship plus the draft margin is less than or equal to the maximum allowable draft, then When the maximum draft is exceeded, it means there is no risk of grounding. -1 indicates that there is a risk of grounding. When there is a risk of collision, deviation or grounding, an alarm will be issued immediately and the position of the risk vessel will be clearly marked on the electronic chart; The mixed formation is controlled based on the evaluation results, and the navigation status of the mixed formation is judged based on the evaluation results. When the navigation status of the mixed formation is abnormal, the alarm information is obtained, and adjustment instructions are issued through the LoRa communication network based on the alarm information. The navigation parameters of the unmanned ship are adjusted based on the adjustment instructions until the risk is eliminated to complete the control of the mixed formation. For example, during the navigation of the mixed formation, once the unmanned ship is detected to deviate from the established course, the driver will quickly send a correction command through the central control platform. Through the LoRa communication hub, these commands are efficiently transmitted to the communication node of the specific unmanned ship. After receiving the command, the unmanned ship will automatically correct the rudder angle and engine speed to quickly restore the heading and speed to ensure the smooth and safe navigation of the formation. The process is: first, plan the track correction. The formation operator carefully plans the track correction plan of the unmanned ship through the integrated control center and formulates detailed adjustment instructions to ensure that the unmanned ship can respond and correct the navigation deviation immediately; transmit the correction instruction, the operator After the operator has verified that everything is correct, the integrated control center uses the LoRa communication hub to send correction instructions to a specific unmanned ship. Relying on the efficient transmission characteristics of LoRa technology, the instructions are ensured to be accurately transmitted to the target ship; interpreting the instruction information, after the unmanned ship's LoRa communication terminal receives the correction instruction, it immediately parses the instruction content, verifies the accuracy of the track and speed adjustment parameters, and prepares for the upcoming navigation adjustment; secondly, performing track adjustment, the unmanned ship automatically adjusts the steering gear system to calibrate the track according to the command analysis results, and synchronously adjusts the engine output to adjust the speed. The coordinated correction of track and speed ensures that the unmanned ship can quickly and smoothly return to the predetermined course, ensuring the order and safety of the fleet's navigation; finally, confirming the adjustment results and feedback: after the track adjustment is completed, the unmanned ship sends a confirmation message of the completion of the adjustment to the integrated control center through the LoRa communication terminal. The integrated control center receives and displays this feedback information to the operator to help him monitor the progress of the adjustment for subsequent navigation planning and decision-making.

[0026] In order to better implement the mixed fleet control method of the present invention, based on the mixed fleet control method of the present invention, correspondingly, Figure 3As shown, the embodiment of the present invention further provides a mixed formation control device for ships, and the mixed formation control device 300 for ships includes: A mixed formation building module 301 is used to build a mixed formation of a manned ship leading an unmanned ship through LoRa technology; The trajectory prediction module 302 is used to obtain dynamic data of the mixed formation, fuse the dynamic data, and then use the EMD-LSTM network to predict the navigation trajectory of the mixed formation based on the fused dynamic data to obtain a predicted trajectory; The mixed formation control module 303 is used to build a navigation risk warning system based on the actual trajectory and predicted trajectory of the mixed formation, evaluate the navigation risk of the mixed formation based on the navigation risk warning system, obtain the evaluation result, and control the mixed formation based on the evaluation result.

[0027] like Figure 4 As shown, the present invention also provides an electronic device 400 , which can be a computing device such as a mobile terminal, a desktop computer, a notebook, a palm computer, a server, etc. The electronic device 400 includes a processor 401 , a memory 402 , and a display 403 . Figure 4 Only some components of the electronic device 400 are shown, but it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0028] In some embodiments, the memory 402 may be an internal storage unit of the electronic device 400, such as a hard disk or memory of the electronic device 400. In other embodiments, the memory 402 may also be an external storage device of the electronic device 400, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 400. Further, the memory 402 may also include both an internal storage unit of the electronic device 400 and an external storage device. The memory 402 is used to store application software and various data installed in the electronic device 400, such as program codes installed in the electronic device 400. The memory 402 may also be used to temporarily store data that has been output or is to be output. In one embodiment, a ship mixed formation control program is stored in the memory 402, and the ship mixed formation control program can be executed by the processor 401, thereby realizing the ship mixed formation control method of each embodiment of the present invention.

[0029] In some embodiments, the processor 401 may be a central processing unit (CPU), a microprocessor or other data processing chip, used to run program codes or process data stored in the memory 402, such as a mixed formation control method for ships.

[0030] In some embodiments, the display 403 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch device, etc. The display 403 is used to display identification information of the ship mixed formation control program and to display a visual user interface. The components 401-403 of the electronic device 400 communicate with each other via a system bus.

[0031] In some embodiments, when the processor 401 executes the ship mixed formation control program in the memory 402, the various steps in the ship mixed formation control method described in the above embodiments are implemented. Since the ship mixed formation control method has been described in detail above, it will not be repeated here.

[0032] Accordingly, the present invention also provides a computer-readable storage medium, which is used to store computer-readable programs or instructions. When the program or instructions are executed by a processor, the steps or functions in the ship mixed formation control method provided in the above-mentioned method embodiments can be implemented.

[0033] In summary, the mixed formation control method, device, equipment and medium provided by the present invention build a mixed formation of manned ships leading unmanned ships through LoRa technology, obtain dynamic data of the mixed formation, fuse the dynamic data, and then use the EMD-LSTM network to predict the navigation trajectory of the mixed formation based on the fused dynamic data to obtain a predicted trajectory, build a navigation risk warning system based on the actual trajectory of the obtained mixed formation and the predicted trajectory, evaluate the navigation risk of the mixed formation based on the navigation risk warning system, obtain an evaluation result, control the mixed formation based on the evaluation result, and improve the efficiency and accuracy of formation decision-making.

[0034] Those skilled in the art will appreciate that all or part of the processes of the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, wherein the computer-readable storage medium is a disk, an optical disk, a read-only storage memory, or a random access memory, etc.

[0035] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed by the present invention should be covered within the protection scope of the present invention.

Claims

1. A mixed fleet control method, characterized in that: include: Use LoRa technology to build a mixed fleet of manned ships leading unmanned ships; acquiring dynamic data of the mixed formation, fusing the dynamic data, and predicting the navigation trajectory of the mixed formation using an EMD-LSTM network based on the fused dynamic data to obtain a predicted trajectory; A navigation risk warning system is constructed based on the actual trajectory of the mixed formation and the predicted trajectory, the navigation risk of the mixed formation is evaluated based on the navigation risk warning system, an evaluation result is obtained, and the mixed formation is controlled based on the evaluation result.

2. The method for controlling a mixed fleet of ships according to claim 1, characterized in that: The hybrid formation of manned ships leading unmanned ships using LoRa technology includes: With manned ships as the LoRa hub and unmanned ships as the LoRa terminals, a mixed fleet of manned ships leading unmanned ships is built.

3. The method for controlling a mixed fleet of ships according to claim 2, characterized in that: The fusing of the dynamic data includes: Acquiring dynamic data of the mixed formation during navigation through AIS, radar, GNSS and electronic compass, wherein the dynamic data at least includes ship position, heading, navigation speed, AIS track and radar track; After refining the dynamic data, refined dynamic data is obtained; The MMSI of the ship is obtained. Based on the refined data, the ship position, heading and AIS are associated through the MMSI of the ship, the AIS is optimized, and the optimized AIS tracks are clustered using DBSCAN. The clustered AIS tracks and radar tracks are bidirectionally aligned using the BDTW algorithm to complete the fusion of dynamic data.

4. The method for controlling a mixed fleet of ships according to claim 3, characterized in that: The EMD-LSTM network includes an EMD algorithm and an LSTM network; the EMD-LSTM network is used to predict the navigation trajectory of the mixed formation based on the fused dynamic data to obtain the predicted trajectory, including: Obtaining historical trajectory characteristics of the ships in the mixed formation based on the fused dynamic data, wherein the historical trajectory characteristics at least include the longitude, latitude, speed and heading of the ships; Inputting the historical trajectory features into the LSTM network, predicting the navigation trajectory of the mixed formation through the LSTM network, and obtaining a first predicted trajectory and a second predicted trajectory; Acquire an actual trajectory of the mixed formation, determine a residual sequence based on the actual trajectory and the first predicted trajectory, decompose the residual sequence by an EMD algorithm, and obtain IMFs and residual terms; Inputting the IMFs and residual items into the LSTM network, predicting the IMFs and residual items through the LSTM network, and obtaining prediction results of the IMFs and residual items; The prediction results of the IMFs and the prediction results of the residual items are superimposed respectively, and a prediction trajectory is obtained based on the second prediction trajectory, the superimposed IMFs and the superimposed residual items.

5. The method for controlling a mixed fleet of ships according to claim 4, characterized in that: The navigation risk warning system includes a collision risk index, a deviation risk index and a grounding risk index; the navigation risk warning system is constructed based on the actual trajectory of the mixed formation and the predicted trajectory, including: Preset a minimum safety interval threshold, determine the positions of the ships in the mixed formation in the future time based on the predicted trajectory, calculate the minimum interval distance between the ships based on the positions of the ships, and determine a first collision risk indicator based on the minimum safety interval threshold and the minimum interval distance, wherein the future time is three minutes; Determining a safety zone for the ships in the mixed formation, determining a location of an obstacle based on the actual trajectory and the predicted trajectory, and determining a second collision risk indicator based on the safety zone for the ships in the mixed formation and the location of the obstacle; determining a collision risk indicator based on the first collision risk indicator and the second collision risk indicator; Preset a safety distance threshold, determine the maximum distance between the unmanned ship and the manned ship based on the actual trajectory and the predicted trajectory, and determine a collision risk index based on the maximum distance between the unmanned ship and the manned ship and the safety distance threshold; The electronic chart information is obtained, and a ship draft margin is preset. The actual draft and maximum draft of the ships in the mixed formation are determined based on the actual trajectory, the predicted trajectory and the electronic chart information. The grounding risk index is determined based on the ship draft margin, the maximum draft and the ship draft margin.

6. The method for controlling a mixed fleet of ships according to claim 5, characterized in that: The calculation formula of the collision risk index is: ， ， ， in, is the collision risk indicator, is the first collision risk indicator, is the second collision risk indicator, The minimum distance between the predicted positions of ships in the fleet within three minutes. is the minimum safety interval threshold, is the location of the obstacle, For the mixed formation The scope of the safety area of ​​the ship; The calculation formula of the deviation risk index is: ， in, is the deviation risk indicator, For the The maximum distance between an unmanned ship and the pilot ship, is the preset safety distance threshold; The calculation formula of the stranding risk index is: ， in, is the stranding risk indicator, For the mixed formation The actual draft of the ship, is the preset draft margin, The maximum draft allowed for the navigation path.

7. The method for controlling a mixed fleet of ships according to claim 5, characterized in that: The controlling the mixed formation based on the evaluation result comprises: The navigation status of the mixed formation is judged based on the evaluation result. When the navigation status of the mixed formation is abnormal, alarm information is obtained. Based on the alarm information, adjustment instructions are issued through the LoRa communication network. Based on the adjustment instructions, the navigation parameters of the unmanned ship are adjusted to complete the control of the mixed formation.

8. A mixed fleet control device, characterized in that: include: A mixed formation building module, which is used to build a mixed formation of manned ships leading unmanned ships through LoRa technology; A trajectory prediction module is used to obtain dynamic data of the mixed formation, fuse the dynamic data, and then predict the navigation trajectory of the mixed formation using an EMD-LSTM network based on the fused dynamic data to obtain a predicted trajectory; A mixed formation control module is used to build a navigation risk warning system based on the actual trajectory of the mixed formation and the predicted trajectory, evaluate the navigation risk of the mixed formation based on the navigation risk warning system, obtain an evaluation result, and control the mixed formation based on the evaluation result.

9. An electronic device, characterized in that: including memory and processor; The memory stores a computer-readable program executable by the processor; When the processor executes the computer-readable program, the steps in the mixed formation control method of ships as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps in the ship mixed formation control method as described in any one of claims 1-7.

Images