Method, device and readable storage medium for determining a ship route

Abstract

The application provides a ship route determination method, device and readable storage medium. The ship route determination method comprises the following steps: obtaining an identification mapping library of a target ship; obtaining a first trajectory data set according to the identification mapping library; performing discrete processing on a navigation parameter in the first trajectory data set to obtain a second trajectory data set with a discrete state label, wherein the navigation parameter at least comprises a heading, a speed and a draft depth; obtaining range data of a target port; obtaining grid data covering the target port according to the range data; obtaining grid heading data and a turning area mark in each grid unit in the grid data according to the second trajectory data set and the grid data; and obtaining a target route according to the grid heading data, the turning area mark and the second trajectory data set. Through a clustering and path search algorithm, a dynamic route conforming to historical optimal practice and suitable for a specific ship state is synthesized.

Description

Technical Field

[0001] This application relates to the field of ship route planning technology, and more specifically, to a method, apparatus and readable storage medium for determining ship routes. Background Technology

[0002] In related technologies, the navigation phase from port anchorage to berth is one of the most complex, traffic-intensive, operationally challenging, and safety-risk stages of a ship's entire voyage. Therefore, route planning is crucial for safety.

[0003] In practice, when obtaining a ship's operating route, it mainly relies on the pilot's personal experience, which is difficult to quantify. Furthermore, tools such as ECDIS (Electronic Chart System) can only provide fixed planned routes and cannot incorporate the ship's dynamic status parameters, resulting in low route accuracy and failure to meet the actual dynamic navigation conditions. Summary of the Invention

[0004] This application aims to address the technical problems existing in the prior art or related technologies, which rely on the pilot's personal experience that is difficult to quantify, and tools such as ECDIS can only provide fixed planned routes and cannot incorporate the dynamic state parameters of the ship, resulting in low route accuracy and inability to meet the actual dynamic navigation conditions.

[0005] Therefore, the first aspect of this application proposes a method for determining ship routes.

[0006] The second aspect of this application proposes a device for determining a ship's route.

[0007] The third aspect of this application proposes a device for determining a ship's route.

[0008] The fourth aspect of this application proposes a readable storage medium.

[0009] In view of this, a method for determining a ship's route is proposed according to the first aspect of this application, comprising: acquiring an identifier mapping library of the target ship; acquiring a first trajectory dataset based on the identifier mapping library; discretizing the navigation parameters in the first trajectory dataset to acquire a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed, and draft; acquiring range data of the target port; acquiring grid data covering the target port based on the range data; acquiring grid heading data and turning area markers within each grid cell of the grid data based on the second trajectory dataset and the grid data; and acquiring the target route based on the grid heading data, the turning area markers, and the second trajectory dataset.

[0010] This application provides a method for determining a ship's route, comprising the following steps: acquiring an identifier mapping library for the target ship; acquiring a first trajectory dataset based on the identifier mapping library; discretizing the navigation parameters in the first trajectory dataset to acquire a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed, and draft; acquiring the range data of the target port; acquiring grid data covering the target port based on the range data; acquiring grid heading data and turning area markers within each grid cell of the grid data based on the second trajectory dataset and the grid data; and acquiring the target route based on the grid heading data, turning area markers, and the second trajectory dataset. The identifier mapping library refers to a database that associates and matches a ship's permanent identification identifier (Maritime Organization number, IMO) with its dynamic communication identifier (Maritime Mobile Service Identifier, MMSI). For example, a ship with an IMO of "1234567" is bound to its MMSI "123456789".

[0011] The first trajectory dataset refers to a collection of raw ship navigation trajectory data extracted from a historical AIS (Historical Navigation Information System) database without in-depth processing. For example, it is a raw data table containing fields such as timestamp, latitude and longitude, heading, speed, and draft.

[0012] The second trajectory dataset refers to a new dataset created by discretizing the first trajectory dataset, with each data point having a discrete state label (heading label, speed level label, and draft level label). For example, an original data point might be labeled as follows: heading label: direction 3, speed label: speed level 5, and draft label: draft level 4.

[0013] Grid data refers to the spatial structure formed by dividing a target port (such as the waters from the anchorage to the berth in XX port) into regular cells (such as 0.01°×0.01° squares).

[0014] Grid heading data refers to the dominant navigation directions obtained through statistical calculations within each grid cell. For example, if statistics show that 80% of the data points in a certain grid have heading labels of "direction 2" and "direction 5", then these two directions are the dominant headings of the grid.

[0015] A turning area marker is an attribute label for a grid cell used to identify key areas where ships frequently change course. For example, the system automatically marks a grid cell with both the main course "Direction 2" (representing the direction of entering the port) and "Direction 5" (representing the direction towards the berth after turning) as a "turning area".

[0016] The target port's waters constitute a continuous geographic space. By dividing the target port into regular grid cells, the spatial region is transformed into statistically significant units, allowing for precise analysis of ship navigation behavior within each grid cell. This gridding of the target port provides a spatial indexing framework for statistical analysis. Furthermore, at the grid cell level, behavioral patterns and key maneuvering nodes of the ship group are mined. Using quantified grid heading data, turning areas, and second trajectory datasets, clustering and path search algorithms are employed to synthesize a dynamic route that conforms to historical best practices and adapts to specific ship conditions.

[0017] This application automatically extracts optimal navigation experience from massive historical navigation data and generates customized safe and efficient routes for ships in different states (draft, speed), thereby significantly improving navigation safety, optimizing port scheduling efficiency, and providing reliable path support for ship autopilot. Specifically, by establishing an identifier mapping library and discretizing the data, the chaotic and continuous raw data is reduced to a structured, labeled dataset, laying the foundation for machine analysis. Then, spatial gridding provides a spatial indexing framework for statistical analysis. Furthermore, at the grid cell level, behavioral patterns (mainstream course) and key maneuvering nodes (turning areas) of the ship group are mined, thereby quantifying implicit human experience. The quantified grid course data, along with turning areas and secondary trajectory datasets, can be comprehensively utilized to synthesize a dynamic route that conforms to historical best practices and adapts to specific ship states through clustering and path search algorithms.

[0018] In some technical solutions, optionally, the identification mapping library of the target vessel is obtained, including: matching the maritime organization number of the target vessel with the maritime mobile communication service identification code according to the vessel file database to obtain the identification mapping library of the target vessel.

[0019] In the technical solution of this application, the ship archive database refers to an authoritative database that records static information of merchant ships worldwide, including ship name, IMO, MMSI, ship type, deadweight tonnage, container capacity, etc.

[0020] Matching refers to the process of programmatically linking the IMO number of a target vessel with the MMSI number it broadcasts during its voyage to ensure data consistency.

[0021] Based on the vessel archive database, the target vessel's Maritime Organization (MMO) number is matched with its Maritime Mobile Service (MMSI) identifier to obtain the target vessel's identifier mapping library. Utilizing the uniqueness and persistence of the IMO number, the data tracking interruption issue caused by potential changes to the MMSI number is resolved. By establishing and relying on a stable IMO-MMSI mapping relationship, the entire historical trajectory of a specific vessel can be accurately located within a large volume of AIS data streams, ensuring the purity of the analysis object.

[0022] In some technical solutions, optionally, a first trajectory dataset is obtained based on an identifier mapping library, including: obtaining historical navigation data of the target vessel, and obtaining multiple navigation trajectory data within a target time interval in the target port area based on the historical navigation data and the identifier mapping library; and obtaining the first trajectory dataset based on the multiple navigation trajectory data.

[0023] In the technical solution of this application, preprocessing refers to the process of cleaning and correcting the raw AIS data to eliminate noise and errors. For example, the program automatically deletes outliers with obvious latitude and longitude on land, or corrects instantaneously abruptly changed airspeed values ​​to the average of previous and subsequent times.

[0024] Data cleaning significantly improves the quality of data relied upon for subsequent analysis. Historical navigation data of the target vessel is acquired, and based on this historical data and an identifier mapping library, multiple navigation trajectory data within the target time interval are obtained from the target port area. Based on these multiple trajectory data, a first trajectory dataset is obtained, and data generated by equipment malfunctions, signal interference, etc., is then removed. This makes subsequent pattern recognition and clustering analysis results more accurate and reliable, improving the accuracy of route planning.

[0025] In some technical solutions, the navigation parameters in the first trajectory dataset can be discretized, including: dividing the heading range from 0° to 360° into multiple sectors, where each sector has a heading label relative to the sector; dividing the speed range into grades at fixed intervals to determine multiple speed grade labels; and dividing the draft range into layers at fixed intervals to determine multiple draft layer labels.

[0026] In the technical solution of this application, a sector refers to a number of angular intervals into which a 360° direction is divided. For example, if it is divided into 8 sectors, each sector covers 45°, such as "Direction 1" representing 0° to 45°, "Direction 2" representing 45° to 90°, and so on.

[0027] Speed ​​rating labels refer to the ratings assigned to different speed levels. For example, with a 2-knot interval, 0 to 2 knots is designated as "Level 1", and 2 to 4 knots is designated as "Level 2".

[0028] Draft level labels refer to the hierarchical identifiers assigned after classifying draft depths. For example, ships with a draft range of 10 to 12 meters are labeled "Level 3" at 2-meter intervals. By mapping an infinite space of continuous values ​​onto a finite set of discrete labels, dimensionality reduction and structuring of data are achieved. This enables computers to efficiently perform statistical analysis and processing of ship behavior, improving the efficiency and accuracy of data processing.

[0029] In some technical solutions, optionally, based on the second trajectory dataset and the grid data, the grid heading data within each grid cell of the grid data is obtained, including: for each grid cell in the grid data, the distribution of heading labels of historical navigation data points falling into each grid cell is statistically analyzed; from the heading label distribution, the top K heading labels with the largest number among multiple historical navigation data points are determined as the mainstream heading of the grid cell; where K is a preset positive integer.

[0030] In the technical solution of this application, the heading label distribution refers to the statistical frequency of heading labels for all AIS data points within a grid cell. For example, if there are 100 points in a grid, there are 50 for "direction 2", 30 for "direction 5", and 20 for the other directions.

[0031] The dominant headings are the K most frequently occurring headings selected from the heading label distribution. For example, if K=2, then "direction 2" and "direction 5" in the example above are determined as the dominant headings of the grid.

[0032] Based on the second trajectory dataset and grid data, obtain the grid heading data within each grid cell in the grid data, including: for each grid cell in the grid data, statistically analyze the heading label distribution of historical navigation data points falling within each grid cell; from the heading label distribution, determine the top K heading labels with the most occurrences among multiple historical navigation data points as the mainstream heading of the grid cell; where K is a preset positive integer, by statistically analyzing the heading choices of all historical ships in a local area, to mine the optimal or safest path that is commonly recognized and followed by the group, thus achieving data extraction and fusion.

[0033] In some technical solutions, optionally, grid cells in which two main headings have a difference greater than a preset threshold are marked as turning areas.

[0034] In the technical solution of this application, a directional difference greater than a preset threshold means that the angle difference between two main headings exceeds a set value. For example, if the threshold is set to 90°, then the grid is considered a turning area only when the angle between two main headings is greater than 90° (such as "direction 1" 0°-45° and "direction 5" 180°-225°).

[0035] Grid cells with two main course directions that have a difference greater than a preset threshold are marked as turning areas. This is based on behavioral pattern mutation detection. In ship navigation, drastic changes in course are the most direct manifestation of maneuvering behavior. Therefore, by detecting two main course directions with huge differences within a small geographical unit, the system can accurately locate the "decision point" where the ship group generally makes course adjustments, which is the "turning point" in the pilot's experience.

[0036] In some technical solutions, optionally, the target route is obtained based on grid heading data, turning area markers, and the second trajectory dataset, including: dividing the second trajectory dataset into multiple navigation condition subsets based on discretized state labels, where each subset corresponds to a combination of heading, speed, and draft; for each subset, using a spatial clustering algorithm to identify the cluster center of the target vessel's position points; determining multiple target waypoints based on the cluster centers; and finding the target route from the starting point to the destination using a path search algorithm based on the target waypoints.

[0037] In the technical solution of this application, the navigation condition subset refers to a subset of data divided according to the discretized state labels, and each subset represents a specific combination of navigation states. For example, "Navigation condition subset A" includes the data points of "heading label: direction 3, speed label: speed level 4, draft label: draft layer 5".

[0038] Spatial clustering algorithms are algorithms used to discover dense clusters of points in a geographic space, such as K-Means or DBSCAN. For example, clustering all latitude and longitude points in "work condition subset A" can identify the core points of three dense regions.

[0039] Cluster centers are points identified by spatial clustering algorithms that represent the core of densely populated areas of ship positions, i.e., key waypoints. For example, the latitude and longitude coordinates of the core points of the three densely populated areas mentioned above.

[0040] Path search algorithms are algorithms used to find the optimal path in a network of nodes (waypoints).

[0041] This application divides the second trajectory dataset into multiple navigation condition subsets based on discretized state labels, where each subset corresponds to a combination of heading, speed, and draft. For each subset, a spatial clustering algorithm is used to identify the cluster centers of the target vessel's position points. Multiple target waypoints are determined based on the cluster centers. A path search algorithm is then used to find the target route from the starting point to the destination based on these waypoints. This enables the generation of distinctly different, customized optimal routes for vessels with different drafts and speeds. This solves the problem of static, fixed, and unadjustable route planning, achieving dynamic and personalized planning and meeting the needs of autonomous driving.

[0042] According to a second aspect of this application, a device for determining a ship's route is provided, comprising: a first acquisition module, a second acquisition module, a third acquisition module, a fourth acquisition module, a fifth acquisition module, a sixth acquisition module, and a seventh acquisition module. The first acquisition module is used to acquire an identifier mapping library of the target ship; the second acquisition module is used to acquire a first trajectory dataset based on the identifier mapping library; the third acquisition module is used to discretize the navigation parameters in the first trajectory dataset to acquire a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed, and draft; the fourth acquisition module is used to acquire range data of the target port; the fifth acquisition module is used to acquire grid data covering the target port based on the range data; the sixth acquisition module is used to acquire grid heading data and turning area markers within each grid cell of the grid data based on the second trajectory dataset and the grid data; and the seventh acquisition module is used to acquire the target route based on the grid heading data, the turning area markers, and the second trajectory dataset.

[0043] The vessel route determination apparatus provided in this application includes: a first acquisition module, a second acquisition module, a third acquisition module, a fourth acquisition module, a fifth acquisition module, a sixth acquisition module, and a seventh acquisition module. The first acquisition module is used to acquire an identifier mapping library of the target vessel; the second acquisition module is used to acquire a first trajectory dataset based on the identifier mapping library; the third acquisition module is used to discretize the navigation parameters in the first trajectory dataset to acquire a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed, and draft; the fourth acquisition module is used to acquire range data of the target port; the fifth acquisition module is used to acquire grid data covering the target port based on the range data; the sixth acquisition module is used to acquire grid heading data and turning area markers within each grid cell of the grid data based on the second trajectory dataset and the grid data; and the seventh acquisition module is used to acquire the target route based on the grid heading data, the turning area markers, and the second trajectory dataset. Dynamic route planning based on the heading angle distribution, speed, and draft of large container ships can solve the problems of inaccurate route planning, poor adaptability, and inability to fully utilize ship dynamic data (heading angle, speed, and draft) for optimization decisions in complex port waters. This can achieve the effects of improving navigation safety, optimizing port scheduling efficiency, enhancing maritime VTS supervision capabilities, and providing reliable path support for autonomous driving, thus promoting the intelligent development of shipping.

[0044] According to a third aspect of this application, a vessel route determination apparatus is provided, comprising: a processor and a memory, wherein the memory stores a program or instructions, and the processor, when executing the program or instructions in the memory, implements the steps of the vessel route determination method as described in any of the above-described technical solutions. Therefore, the vessel route determination apparatus possesses all the beneficial effects of the vessel route determination method as described in any of the above-described technical solutions.

[0045] According to the fourth aspect of this application, a readable storage medium is provided, on which a program or instructions are stored, which, when executed by a processor, implement the steps of the method for determining a ship's route as described in any of the above-described technical solutions. Therefore, the readable storage medium possesses all the beneficial effects of the method for determining a ship's route as described in any of the above-described technical solutions.

[0046] Additional aspects and advantages of this application will become apparent in the following description or may be learned by practice of this application. Attached Figure Description

[0047] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which:

[0048] Figure 1 A flowchart of a method for determining a ship route provided in some embodiments of this application is shown;

[0049] Figure 2 The following is a structural block diagram of one of the embodiments of a device for determining a ship's route provided in this application;

[0050] Figure 3 The second structural block diagram of a device for determining a ship's route is shown in some embodiments of this application;

[0051] Figure 4 The following diagram illustrates the course discretization provided in some embodiments of this application;

[0052] Figure 5 Ship trajectory diagrams provided in some embodiments of this application are shown;

[0053] Figure 6 A schematic diagram of the flight path provided in some embodiments of this application is shown. Detailed Implementation

[0054] To better understand the above-mentioned objectives, features, and advantages of this application, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, these embodiments and the features described herein can be combined with each other.

[0055] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, this application may also be implemented in other ways different from those described herein. Therefore, the scope of protection of this application is not limited to the specific embodiments disclosed below.

[0056] The following reference Figures 1 to 6 This application describes a method, apparatus, and readable storage medium for determining ship routes according to some embodiments.

[0057] In one embodiment of this application, such as Figure 1 As shown, a method for determining a ship's route is provided, the steps of which include:

[0058] Step 102: Obtain the identifier mapping library of the target vessel;

[0059] Step 104: Obtain the first trajectory dataset based on the identifier mapping library;

[0060] Step 106: Discretize the navigation parameters in the first trajectory dataset to obtain a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed and draft.

[0061] Step 108: Obtain the range data of the target port;

[0062] Step 110: Based on the range data, obtain the grid data covering the target port;

[0063] Step 112: Based on the second trajectory dataset and grid data, obtain the grid heading data and turning area markers within each grid cell of the grid data;

[0064] Step 114: Obtain the target route based on the grid heading data, turning area markers, and the second trajectory dataset.

[0065] This application provides a method for determining a ship's route, comprising the following steps: acquiring an identifier mapping library for the target ship; acquiring a first trajectory dataset based on the identifier mapping library; discretizing the navigation parameters in the first trajectory dataset to acquire a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed, and draft; acquiring the range data of the target port; acquiring grid data covering the target port based on the range data; acquiring grid heading data and turning area markers within each grid cell of the grid data based on the second trajectory dataset and the grid data; and acquiring the target route based on the grid heading data, turning area markers, and the second trajectory dataset. The identifier mapping library refers to a database that associates and matches a ship's permanent identification identifier (Maritime Organization number, IMO) with its dynamic communication identifier (Maritime Mobile Service Identifier, MMSI). For example, a ship with an IMO of "1234567" is bound to its MMSI "123456789".

[0066] The first trajectory dataset refers to a collection of raw ship navigation trajectory data extracted from historical AIS databases without in-depth processing. For example, it may be a raw data table containing fields such as timestamp, latitude and longitude, heading, speed, and draft.

[0067] The second trajectory dataset refers to a new dataset created by discretizing the first trajectory dataset, with each data point having a discrete state label (heading label, speed level label, and draft level label). For example, an original data point might be labeled as follows: heading label: direction 3, speed label: speed level 5, and draft label: draft level 4.

[0068] Grid data refers to the spatial structure formed by dividing a target port (such as the waters from the anchorage to the berth in Tianjin Port) into regular cells (such as 0.01°×0.01° squares).

[0069] Grid heading data refers to the dominant navigation directions obtained through statistical calculations within each grid cell. For example, if statistics show that 80% of the data points in a certain grid have heading labels of "direction 2" and "direction 5", then these two directions are the dominant headings of the grid.

[0070] A turning area marker is an attribute label for a grid cell used to identify key areas where ships frequently change course. For example, the system automatically marks a grid cell with both the main course "Direction 2" (representing the direction of entering the port) and "Direction 5" (representing the direction towards the berth after turning) as a "turning area".

[0071] The target port's waters constitute a continuous geographic space. By dividing the target port into regular grid cells, the spatial region is transformed into statistically quantifiable units, allowing for precise analysis of ship navigation behavior within each grid cell. This spatial gridding of the target port provides a spatial indexing framework for statistical analysis. Furthermore, at the grid cell level, behavioral patterns and key maneuvering nodes of the ship group are mined. Using quantified grid heading data, turning areas, and second trajectory datasets, clustering and path search algorithms are employed to synthesize a dynamic route that conforms to historical best practices and adapts to specific ship states. This application automatically extracts optimal navigation experience from massive amounts of historical navigation data and generates customized safe and efficient routes for ships in different states (draft, speed), thereby significantly improving navigation safety, optimizing port scheduling efficiency, and providing reliable path support for ship autopilot. Specifically, by establishing an identifier mapping library and discretizing the data, the chaotic and continuous raw data is reduced to a structured, labeled dataset, laying the foundation for machine analysis. Then, spatial gridding provides a spatial indexing framework for statistical analysis. Next, behavioral patterns (mainstream course) and key maneuvering nodes (turning areas) of the ship group are mined at the grid cell level, thereby quantifying implicit human experience. The quantified grid course data, along with turning areas and second trajectory datasets, can be comprehensively utilized to synthesize a dynamic route that conforms to historical best practices and adapts to specific ship conditions through clustering and path search algorithms.

[0072] In some technical solutions, optionally, the identification mapping library of the target vessel is obtained, including: matching the maritime organization number of the target vessel with the maritime mobile communication service identification code according to the vessel file database to obtain the identification mapping library of the target vessel.

[0073] In the technical solution of this application, the ship archive database refers to an authoritative database that records static information of merchant ships worldwide, including ship name, IMO, MMSI, ship type, deadweight tonnage, container capacity, etc.

[0074] Matching refers to the process of programmatically linking the IMO number of a target vessel with the MMSI number it broadcasts during its voyage to ensure data consistency.

[0075] Based on the vessel archive database, the target vessel's Maritime Organization (MMO) number is matched with its Maritime Mobile Service (MMSI) identifier to obtain the target vessel's identifier mapping library. Utilizing the uniqueness and persistence of the IMO number, the data tracking interruption issue caused by potential changes to the MMSI number is resolved. By establishing and relying on a stable IMO-MMSI mapping relationship, the entire historical trajectory of a specific vessel can be accurately located within a large volume of AIS data streams, ensuring the purity of the analysis object.

[0076] In some technical solutions, optionally, a first trajectory dataset is obtained based on an identifier mapping library, including: obtaining historical navigation data of the target vessel, and obtaining multiple navigation trajectory data within a target time interval in the target port area based on the historical navigation data and the identifier mapping library; and obtaining the first trajectory dataset based on the multiple navigation trajectory data.

[0077] In the technical solution of this application, preprocessing refers to the process of cleaning and correcting the raw AIS data to eliminate noise and errors. For example, the program automatically deletes outliers with obvious latitude and longitude on land, or corrects instantaneously abruptly changed airspeed values ​​to the average of previous and subsequent times.

[0078] Data cleaning significantly improves the quality of data relied upon for subsequent analysis. Historical navigation data of the target vessel is acquired, and based on this historical data and an identifier mapping library, multiple navigation trajectory data within the target time interval are obtained from the target port area. Based on these multiple trajectory data, a first trajectory dataset is obtained, and data generated by equipment malfunctions, signal interference, etc., is then removed. This makes subsequent pattern recognition and clustering analysis results more accurate and reliable, improving the accuracy of route planning.

[0079] In some technical solutions, the navigation parameters in the first trajectory dataset can be discretized, including: dividing the heading range from 0° to 360° into multiple sectors, where each sector has a heading label relative to the sector; dividing the speed range into grades at fixed intervals to determine multiple speed grade labels; and dividing the draft range into layers at fixed intervals to determine multiple draft layer labels.

[0080] In the technical solution of this application, a sector refers to a number of angular intervals into which a 360° direction is divided. For example, if it is divided into 8 sectors, each sector covers 45°, such as "Direction 1" representing 0° to 45°, "Direction 2" representing 45° to 90°, and so on.

[0081] Speed ​​rating labels refer to the ratings assigned to different speed levels. For example, with a 2-knot interval, 0 to 2 knots is designated as "Level 1", and 2 to 4 knots is designated as "Level 2".

[0082] Draft level labels refer to the hierarchical identifiers assigned after classifying draft depths. For example, ships with a draft range of 10 to 12 meters are labeled "Level 3" at 2-meter intervals. By mapping an infinite space of continuous values ​​onto a finite set of discrete labels, dimensionality reduction and structuring of data are achieved. This enables computers to efficiently perform statistical analysis and processing of ship behavior, improving the efficiency and accuracy of data processing.

[0083] In some technical solutions, optionally, based on the second trajectory dataset and the grid data, the grid heading data within each grid cell of the grid data is obtained, including: for each grid cell in the grid data, the distribution of heading labels of historical navigation data points falling into each grid cell is statistically analyzed; from the heading label distribution, the top K heading labels with the largest number among multiple historical navigation data points are determined as the mainstream heading of the grid cell; where K is a preset positive integer.

[0084] In the technical solution of this application, the heading label distribution refers to the statistical frequency of heading labels for all AIS data points within a grid cell. For example, if there are 100 points in a grid, there are 50 for "direction 2", 30 for "direction 5", and 20 for the other directions.

[0085] The dominant headings are the K most frequently occurring headings selected from the heading label distribution. For example, if K=2, then "direction 2" and "direction 5" in the example above are determined as the dominant headings of the grid.

[0086] Based on the second trajectory dataset and grid data, obtain the grid heading data within each grid cell in the grid data, including: for each grid cell in the grid data, statistically analyze the heading label distribution of historical navigation data points falling within each grid cell; from the heading label distribution, determine the top K heading labels with the most occurrences among multiple historical navigation data points as the mainstream heading of the grid cell; where K is a preset positive integer, by statistically analyzing the heading choices of all historical ships in a local area, to mine the optimal or safest path that is commonly recognized and followed by the group, thus achieving data extraction and fusion.

[0087] In some technical solutions, optionally, grid cells in which two main headings have a difference greater than a preset threshold are marked as turning areas.

[0088] In the technical solution of this application, a directional difference greater than a preset threshold means that the angle difference between two main headings exceeds a set value. For example, if the threshold is set to 90°, then the grid is considered a turning area only when the angle between two main headings is greater than 90° (such as "direction 1" 0°-45° and "direction 5" 180°-225°).

[0089] Grid cells with two main course directions that have a difference greater than a preset threshold are marked as turning areas. This is based on behavioral pattern mutation detection. In ship navigation, drastic changes in course are the most direct manifestation of maneuvering behavior. Therefore, by detecting two main course directions with huge differences within a small geographical unit, the system can accurately locate the "decision point" where the ship group generally makes course adjustments, which is the "turning point" in the pilot's experience.

[0090] In some technical solutions, optionally, the target route is obtained based on grid heading data, turning area markers, and the second trajectory dataset, including: dividing the second trajectory dataset into multiple navigation condition subsets based on discretized state labels, where each subset corresponds to a combination of heading, speed, and draft; for each subset, using a spatial clustering algorithm to identify the cluster center of the target vessel's position points; determining multiple target waypoints based on the cluster centers; and finding the target route from the starting point to the destination using a path search algorithm based on the target waypoints.

[0091] In the technical solution of this application, the navigation condition subset refers to a subset of data divided according to the discretized state labels, and each subset represents a specific combination of navigation states. For example, "Navigation condition subset A" includes the data points of "heading label: direction 3, speed label: speed level 4, draft label: draft layer 5".

[0092] Spatial clustering algorithms are algorithms used to discover dense clusters of points in a geographic space, such as K-Means or DBSCAN. For example, clustering all latitude and longitude points in "work condition subset A" can identify the core points of three dense regions.

[0093] Cluster centers are points identified by spatial clustering algorithms that represent the core of densely populated areas of ship positions, i.e., key waypoints. For example, the latitude and longitude coordinates of the core points of the three densely populated areas mentioned above.

[0094] Path search algorithms are algorithms used to find the optimal path in a network of nodes (waypoints).

[0095] This application divides the second trajectory dataset into multiple navigation condition subsets based on discretized state labels, where each subset corresponds to a combination of heading, speed, and draft. For each subset, a spatial clustering algorithm is used to identify the cluster centers of the target vessel's position points. Multiple target waypoints are determined based on the cluster centers. A path search algorithm is then used to find the target route from the starting point to the destination based on these waypoints. This enables the generation of distinctly different, customized optimal routes for vessels with different drafts and speeds. This solves the problem of static, fixed, and unadjustable route planning, achieving dynamic and personalized planning and meeting the needs of autonomous driving.

[0096] In one embodiment of this application, such as Figure 2As shown, a vessel route determination device 200 is provided, comprising: a first acquisition module 202, a second acquisition module 204, a third acquisition module 206, a fourth acquisition module 208, a fifth acquisition module 210, a sixth acquisition module 212, and a seventh acquisition module 214. The first acquisition module 202 is used to acquire an identifier mapping library of the target vessel; the second acquisition module 204 is used to acquire a first trajectory dataset based on the identifier mapping library; the third acquisition module 206 is used to discretize the navigation parameters in the first trajectory dataset to acquire a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed, and draft; the fourth acquisition module 208 is used to acquire range data of the target port; the fifth acquisition module 210 is used to acquire grid data covering the target port based on the range data; the sixth acquisition module 212 is used to acquire grid heading data and turning area markers within each grid cell of the grid data based on the second trajectory dataset and the grid data; and the seventh acquisition module 214 is used to acquire the target route based on the grid heading data, the turning area markers, and the second trajectory dataset.

[0097] The vessel route determination device 200 provided in this application includes: a first acquisition module 202, a second acquisition module 204, a third acquisition module 206, a fourth acquisition module 208, a fifth acquisition module 210, a sixth acquisition module 212, and a seventh acquisition module 214. The first acquisition module 202 is used to acquire an identifier mapping library of the target vessel; the second acquisition module 204 is used to acquire a first trajectory dataset based on the identifier mapping library; the third acquisition module 206 is used to discretize the navigation parameters in the first trajectory dataset to acquire a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed, and draft; the fourth acquisition module 208 is used to acquire range data of the target port; the fifth acquisition module 210 is used to acquire grid data covering the target port based on the range data; the sixth acquisition module 212 is used to acquire grid heading data and turning area markers within each grid cell of the grid data based on the second trajectory dataset and the grid data; and the seventh acquisition module 214 is used to acquire the target route based on the grid heading data, the turning area markers, and the second trajectory dataset. Dynamic route planning based on the heading angle distribution, speed, and draft of large container ships can solve the problems of inaccurate route planning, poor adaptability, and inability to fully utilize ship dynamic data (heading angle, speed, and draft) for optimization decisions in complex port waters. This can achieve the effects of improving navigation safety, optimizing port scheduling efficiency, enhancing maritime VTS supervision capabilities, and providing reliable path support for autonomous driving, thus promoting the intelligent development of shipping.

[0098] In one embodiment of this application, such as Figure 3As shown, a ship route determination apparatus 300 is provided, including a processor 302 and a memory 304. The memory 304 stores a program or instructions. When the processor 302 executes the program or instructions in the memory 304, it implements the steps of the ship route determination method as described in any of the above embodiments. Therefore, the ship route determination apparatus 300 possesses all the beneficial effects of the ship route determination method as described in any of the above embodiments.

[0099] In one embodiment of this application, a readable storage medium is provided, on which a program or instructions are stored. When executed by a processor, the program or instructions implement the steps of the ship route determination method as described in any of the above embodiments. Therefore, the readable storage medium possesses all the beneficial effects of the ship route determination method as described in any of the above embodiments.

[0100] Specifically, this embodiment selects large container ships with a capacity (TEU) greater than 12,500 TEU from the ship archive database as the target research object. After setting the data retrieval and screening conditions, 762 ships meeting the conditions were obtained from the database, and the Maritime Organization number (IMO) and Maritime Mobile Services Identifier (MMSI) of each ship were precisely matched to establish a target ship identification information database.

[0101] Based on this, all dynamic navigation trajectory data points of the aforementioned 762 ships in the XX port area for three consecutive months were extracted from the historical AIS data center, with a total data volume of approximately 960,000 records.

[0102] Based on predefined ship type and container capacity conditions, the target vessel is identified, and a one-to-one correspondence is established between the IMO and MMSI, forming a target vessel identification table. Table 1 below serves as the unique index for subsequent data extraction and matching, ensuring that the extracted trajectory data accurately corresponds to the target vessel.

[0103] Historical navigation data of the target vessel in the Tianjin Port area was retrieved from the AIS data center, and a series of data quality control and preprocessing operations were performed, including: removing or correcting abnormal geographical locations caused by signal drift, positioning anomalies or equipment failures; deleting AIS data with a course change of more than ±90° that does not conform to actual dynamics; removing illogical speed values ​​(such as zero speed but continuous position changes); and correcting draft values ​​that are inconsistent with the navigation status (such as full load draft appearing in an empty state).

[0104] The preprocessed AIS dynamic data underwent three-dimensional discretization, including heading discretization: the range from 0° to 360° was divided into 8 directional intervals of 45° each, assigned directional labels H1–H8; speed discretization: 0–20 knots were divided into speed increments of 2 knots each, forming speed increments S1–S10; draft discretization: the range from 8–20 meters was divided into draft increments of 2 meters each, forming draft steps D1–D6. Each AIS data point was assigned a three-dimensional condition label (Ci, Sj, Dk), resulting in approximately 118 valid combinations (some combinations did not appear in the actual data).

[0105] Table 1

[0106]

[0107] Specifically, such as Figure 4 As shown in the figure, region A represents the discretized 360° heading distribution.

[0108] Specifically, such as Figure 5 and Figure 6 As shown, Figure 5 The trajectory shown in B is the original trajectory without preprocessing. Figure 6 The diagram shows the planned navigation key points: WayPoint1, WayPoint4, WayPoint7, and WayPoint9. Routes can be planned based on these multiple navigation key points.

[0109] It should be clarified that in the claims, description, and accompanying drawings of this application, the term "multiple" refers to two or more objects. Unless otherwise explicitly defined, the terms "upper," "lower," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing this application and simplifying the description process, not to indicate or imply that the device or element referred to must have the described specific orientation, or be constructed and operated in a specific orientation. Therefore, these descriptions should not be construed as limitations on this application. The terms "connection," "installation," "fixing," etc., should be interpreted broadly. For example, "connection" can be a fixed connection between multiple objects, a detachable connection between multiple objects, or an integral connection; it can be a direct connection between multiple objects or an indirect connection between multiple objects through an intermediate medium. For those skilled in the art, the specific meaning of the above terms in this application can be understood based on the specific circumstances of the above data.

[0110] In the claims, description, and accompanying drawings of this application, the terms "one embodiment," "some embodiments," "specific embodiment," etc., refer to a specific feature, structure, material, or characteristic described in connection with that embodiment or example, which is included in at least one embodiment or example of this application. In the claims, description, and accompanying drawings of this application, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0111] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for determining a ship's route, characterized in that, include: Obtain the identifier mapping library of the target vessel; According to the identifier mapping library, obtain the first trajectory dataset; The navigation parameters in the first trajectory dataset are discretized to obtain a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed and draft. Obtain the range data of the target port; Based on the range data, obtain grid data covering the target port; Based on the second trajectory dataset and the grid data, obtain the grid heading data and turning area markers within each grid cell of the grid data; The target route is obtained based on the grid heading data, the turning area markers, and the second trajectory dataset; The step of obtaining the target route based on the grid heading data, the turning area markers, and the second trajectory dataset includes: Based on the discretized state labels, the second trajectory dataset is divided into multiple navigation condition subsets, where each subset corresponds to a combination of heading, speed and draft. For each subset, a spatial clustering algorithm is used to identify the cluster centers of the target vessel's position points; Multiple target waypoints are determined based on the cluster centers; Based on the target waypoint, a path search algorithm is used to find the target route from the origin to the destination.

2. The method for determining a ship's route according to claim 1, characterized in that, The acquisition of the target vessel's identifier mapping library includes: Based on the ship file database, the maritime organization number of the target ship is matched with the maritime mobile communication service identification code to obtain the identification mapping library of the target ship.

3. The method for determining a ship's route according to claim 1, characterized in that, The step of obtaining the first trajectory dataset according to the identifier mapping library includes: The historical navigation data of the target vessel is obtained, and based on the historical navigation data and the identifier mapping library, multiple navigation trajectory data within the target time interval in the target port area are obtained. Based on multiple navigation trajectory data, a first trajectory dataset is obtained.

4. The method for determining a ship's route according to claim 1, characterized in that, The discretization of the navigation parameters in the first trajectory dataset includes: The heading range from 0° to 360° is divided into multiple sectors, wherein each sector has a heading label opposite to the sector; The speed range is divided into fixed intervals, and multiple speed level labels are defined. The draft depth range is divided into layers at fixed intervals, and multiple draft level labels are defined.

5. The method for determining a ship's route according to claim 4, characterized in that, The step of obtaining the grid heading data within each grid cell of the grid data based on the second trajectory dataset and the grid data includes: For each grid cell in the grid data, the distribution of heading labels for historical navigation data points falling within each grid cell is statistically analyzed. From the distribution of heading labels, the top K heading labels with the most occurrences among the various historical navigation data points are determined as the mainstream heading of the grid cell; Where K is a preset positive integer.

6. The method for determining a ship's route according to claim 5, characterized in that, Also includes: Grid cells in the mainstream headings where the difference between two directions is greater than a preset threshold are identified as turning areas.

7. A device for determining a ship's route, characterized in that, include: The first acquisition module is used to acquire the identifier mapping library of the target vessel; The second acquisition module is used to acquire the first trajectory dataset according to the identifier mapping library; The third acquisition module is used to discretize the navigation parameters in the first trajectory dataset to obtain a second trajectory dataset with discretized state labels, wherein the navigation parameters include at least heading, speed and draft. The fourth acquisition module is used to acquire the range data of the target port; The fifth acquisition module is used to acquire grid data covering the target port based on the range data; The sixth acquisition module is used to acquire, based on the second trajectory dataset and the grid data, the grid heading data and turning area markers within each grid cell of the grid data; The seventh acquisition module is used to acquire the target route based on the grid heading data, the turning area markers, and the second trajectory dataset; The step of obtaining the target route based on the grid heading data, the turning area markers, and the second trajectory dataset includes: Based on the discretized state labels, the second trajectory dataset is divided into multiple navigation condition subsets, where each subset corresponds to a combination of heading, speed and draft. For each subset, a spatial clustering algorithm is used to identify the cluster centers of the target vessel's position points; Multiple target waypoints are determined based on the cluster centers; Based on the target waypoint, a path search algorithm is used to find the target route from the origin to the destination.

8. A device for determining a ship's route, characterized in that, include: processor; A memory storing programs or instructions, wherein the processor, when executing the programs or instructions in the memory, implements the steps of the method for determining a ship's route as described in any one of claims 1 to 6.

9. A readable storage medium, characterized in that, The readable storage medium stores a program or instructions that, when executed by a processor, implement the steps of the method for determining a ship's route as described in any one of claims 1 to 6.

Images