Method and device for route reconstruction among multiple ports, electronic equipment, readable storage medium and chip

Abstract

Embodiments of the present application provide a route reconstruction method and device among multiple ports, electronic equipment, readable storage medium and chip, wherein the route reconstruction method comprises: obtaining a preset port sequence and a port event database; filtering a sample ship list meeting the port sequence within a set observation time period from the port event database; extracting ship automatic identification system data of at least one sample ship from a starting port to a destination port based on the sample ship list; processing the ship automatic identification system data to identify multiple sailing trajectory segments; fusing and reconstructing the multiple sailing trajectory segments to determine a continuous route from the starting port to the destination port; and determining a route reconstruction path from the starting port to the destination port according to the multiple continuous routes. Through the scheme of the present application, the route calculation accuracy is enhanced, and the intelligentization and scientific level of shipping management are improved.

Description

Technical Field

[0001] This invention relates to the field of ship data technology, and more specifically, to a method, apparatus, electronic device, readable storage medium, and chip for reconstructing routes between multiple ports. Background Technology

[0002] In recent years, with the continuous growth of global maritime transport and the deepening expansion of international trade networks, the digital and intelligent transformation of the shipping industry has become a core trend in industry development. In the context of globalization, the planning and economic assessment of long-distance transcontinental shipping routes play a crucial role in shipping companies' operational decisions, cost control, and logistics efficiency improvement.

[0003] In actual navigation operations, a ship's planned transport route is usually from the port of origin directly to the port of destination. However, due to reasons such as refueling stops, crew replenishment, equipment maintenance, and waiting for navigation, the ship's actual navigation path is often naturally divided into multiple segments. Viewing only the trajectory of a single ship cannot provide a clear, intuitive, and logical representation of long-distance transcontinental shipping routes. Summary of the Invention

[0004] The purpose of this invention is to provide a method, apparatus, electronic device, readable storage medium, and chip for reconstructing routes between multiple ports, which can solve the problem of not being able to construct a complete operating path for intercontinental routes on a global scale.

[0005] In view of this, an embodiment of the first aspect of the present invention provides a method for reconstructing routes between multiple ports.

[0006] A second aspect of the present invention provides a route reconfiguration device between multiple ports.

[0007] An embodiment of the third aspect of the present invention provides an electronic device.

[0008] An embodiment of the fourth aspect of the present invention provides a readable storage medium.

[0009] An embodiment of the fifth aspect of the present invention provides a chip.

[0010] To achieve the above objectives, an embodiment of the first aspect of the present invention provides a method for reconstructing shipping routes between multiple ports, comprising: acquiring a preset port sequence and a port event database; filtering a list of sample vessels conforming to the port sequence within a set observation period from the port event database, wherein the port sequence includes a port of origin, at least one intermediate port, and a port of destination; extracting Automatic Identification System (AIS) data of at least one sample vessel from the port of origin to the port of destination based on the list of sample vessels; processing the AIS data to identify multiple navigation trajectory segments of the sample vessel between the port of origin, at least one intermediate port, and the port of destination; fusing and reconstructing the multiple navigation trajectory segments to determine a continuous shipping route from the port of origin to the port of destination; determining a reconstructed shipping route path from the port of origin to the port of destination based on the multiple continuous shipping routes, and associating and storing the reconstructed shipping route path with the attribute information of at least one sample vessel; wherein the at least one intermediate port passed through by the multiple continuous shipping routes in the reconstructed shipping route path is different.

[0011] The multi-port route reconstruction method provided by this invention selects eligible sample vessels and their historical Automatic Identification System (AIS) data from a port event database using a preset sequence including the origin port, at least one intermediate port, and the destination port. Then, by processing the AIS data, the method identifies the navigation trajectory segments of the sample vessels between multiple ports and merges these segments to reconstruct a continuous route from the origin to the destination. Finally, these continuous routes are associated with and stored with vessel attribute information, thereby constructing a dynamic and refined route reconstruction path. This reconstructed route includes multiple continuous routes passing through different intermediate ports, rather than a single static path.

[0012] In some technical solutions, optionally, the data from the Automatic Identification System (AIS) is processed to identify multiple navigation trajectory segments of a sample vessel between its port of origin, at least one intermediate port, and its port of destination. This includes: performing outlier detection and cleaning on the AIS data to determine preprocessed data; constructing a main channel corridor connecting the port of origin and the port of destination based on the preprocessed data; determining at least one trajectory point corresponding to a single sample vessel; determining the single vessel trajectory based on at least one trajectory point falling within the range corresponding to the main channel corridor; obtaining a first vessel speed threshold; and determining multiple navigation trajectory segments corresponding to the single vessel trajectory based on the first vessel speed threshold.

[0013] In this scheme, the AIS data of the sample vessels is first cleaned to ensure data quality. Then, based on the trajectory points of multiple sample vessels, spatial clustering is used to identify areas with high point density, forming a strip-shaped main track corridor. The main track corridor reflects the historical main co-current channels of the sample vessels on a specific route. For a single sample vessel, each trajectory point is traversed to check whether the trajectory point falls within the range of the main track corridor. If the trajectory point is not within the corridor, it is considered an anomaly and removed. Based on at least one trajectory point falling within the corresponding range of the main track corridor, the individual trajectory of the sample vessel is determined.

[0014] In some technical solutions, optionally, multiple navigation trajectory segments corresponding to a single ship's trajectory are determined based on a first ship speed threshold, including: obtaining speed parameters corresponding to trajectory points; determining at least one low-speed segment based on the speed parameters and the first ship speed threshold, wherein the speed parameters of trajectory points in the low-speed segment are less than the first ship speed threshold; determining the trajectory start point and trajectory end point of the single ship's trajectory based on the low-speed segment; and determining multiple navigation trajectory segments from the trajectory start point and trajectory end point based on trajectory points whose speed parameters are less than the first ship speed threshold.

[0015] In this scheme, a preset first ship speed threshold is used as a judgment standard to automatically identify the low-speed segment in the trajectory segment in order to accurately locate the starting point and ending point of the trajectory in the sample ship's journey, and to divide the single ship trajectory into multiple navigation trajectory segments according to the first ship speed threshold.

[0016] In some technical solutions, optionally, multiple navigation trajectory segments are merged and reconstructed to determine a continuous route from the port of origin to the port of destination. This includes performing the following operations on the navigation trajectory segment between two consecutive ports: obtaining a second ship speed threshold, which is greater than a first ship speed threshold; removing trajectory points in the navigation trajectory segment whose speed is lower than the second ship speed threshold; for trajectory breaks caused by removal, obtaining adjacent trajectory point pairs; obtaining the distance parameter and azimuth change parameter of the adjacent trajectory point pairs; obtaining a first preset tolerance threshold and a second preset tolerance threshold; connecting adjacent trajectory point pairs when the distance parameter is less than or equal to the first preset tolerance threshold and the azimuth change parameter is less than or equal to the second preset tolerance threshold; and completing the connection by performing path completion between adjacent trajectory point pairs based on the main navigation corridor when the distance parameter is greater than the first preset tolerance threshold or the azimuth change parameter is greater than the second preset tolerance threshold.

[0017] In this scheme, the navigation trajectory segments of sample ships are fused and reconstructed based on speed thresholds and geometric features. A second ship speed threshold, greater than the first, is used to actively eliminate low-speed navigation trajectories near ports. Then, the rationality of connection points before and after the eliminated trajectory breaks is determined: if the trajectories on both sides of the break are continuous in distance and heading, they are directly connected; if not, based on the main channel corridor, an algorithm is used within a local area to fill in the connection points before and after the eliminated trajectory breaks. This seamlessly reconstructs multiple navigation trajectory segments interrupted by ports into a continuous route from the origin port to the destination port.

[0018] In some technical solutions, optionally, determining a continuous route from the port of origin to the port of destination includes: obtaining timing parameters of the port sequence; determining the sailing order of at least one intermediate port based on the timing parameters; and determining a continuous route from the port of origin to the port of destination based on the sailing order.

[0019] In this scheme, the time sequence parameter refers to the precise time information associated with the initial port, at least one intermediate port, and the destination port in the port sequence. The time sequence parameter is derived from the berthing and departure timestamps of each sample vessel at each port, recorded in the port event database. Determining the time sequence parameter using the berthing and departure timestamps to judge the sailing order at at least one intermediate port improves the realism and objectivity of the route reconstruction.

[0020] In some technical solutions, optionally, the route reconstruction path is associated with and stored with the attribute information of at least one sample vessel, including: obtaining the timestamp from the Automatic Identification System (AIS) data; determining the actual sailing time of the sample vessel based on the timestamp; obtaining at least one vessel registration information corresponding to the sample vessel list; determining the basic attributes of the sample vessel based on the vessel registration information and the port event database; determining the attribute information of each consecutive route in the route reconstruction path based on the basic attributes and the actual sailing time; and associating and storing the consecutive routes with the attribute information of the corresponding vessels.

[0021] In this solution, by extracting and integrating the timestamps from the Automatic Identification System (AIS) data and the static attributes from the ship registration information, attribute information containing the dimensions of sailing time and the characteristics of the ship itself is generated. The attribute information is then accurately associated with and stored with the reconstructed continuous routes, thereby upgrading a single geometric route into a digital route rich in semantics that supports multi-dimensional queries and refined analysis.

[0022] A second aspect of the present invention provides a multi-port route reconstruction device, comprising: a data acquisition module for acquiring a preset port sequence and a port event database; a sample identification module for filtering a list of sample vessels conforming to the port sequence within a set observation period from the port event database, the port sequence including a port of origin, at least one intermediate port, and a port of destination; a data extraction module for extracting Automatic Identification System (AIS) data of at least one sample vessel from the port of origin to the port of destination based on the list of sample vessels; a trajectory identification module for processing the AIS data to identify multiple navigation trajectory segments of the sample vessels between the port of origin, at least one intermediate port, and the port of destination; a trajectory fusion module for fusing and reconstructing the multiple navigation trajectory segments to determine a continuous route from the port of origin to the port of destination; and a route reconstruction module for determining a route reconstruction path from the port of origin to the port of destination based on the multiple continuous routes, and associating and storing the route reconstruction path with the attribute information of at least one sample vessel; wherein the at least one intermediate port passed through by the multiple continuous routes in the route reconstruction path is different.

[0023] An embodiment of the third aspect of this application provides an electronic device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the route reconstruction method as described in the first aspect.

[0024] An embodiment of the fourth aspect of this application provides a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the route reconstruction method as described in the first aspect.

[0025] An embodiment of the fifth aspect of this application provides a chip including a processor and a communication interface, the communication interface and the processor being coupled together, the processor being used to run a program or instructions to implement the steps of the route reconstruction method as described in the first aspect.

[0026] Additional aspects and advantages of the technical solutions of the present invention will become apparent in the following description or may be learned by practice of the invention. Attached Figure Description

[0027] Figure 1 One of the flowcharts of the multi-port route reconstruction method according to this application is shown;

[0028] Figure 2 A second flowchart illustrating the multi-port route reconstruction method according to this application is shown.

[0029] Figure 3The third flowchart illustrates the multi-port route reconstruction method according to this application;

[0030] Figure 4 The fourth flowchart illustrates the multi-port route reconstruction method according to this application;

[0031] Figure 5 The fifth flowchart illustrates the multi-port route reconstruction method according to this application;

[0032] Figure 6 The sixth flowchart illustrates the multi-port route reconstruction method according to this application;

[0033] Figure 7 A schematic block diagram of the multi-port route reconfiguration device according to this application is shown;

[0034] Figure 8 A schematic block diagram of the structure of an electronic device according to this application is shown.

[0035] Among them, 900: multi-port route reconstruction device; 902: data acquisition module; 904: sample recognition module; 906: data extraction module; 908: trajectory recognition module; 910: trajectory fusion module; 912: route reconstruction module; 1000: electronic device; 1109: memory; 1110: processor. Detailed Implementation

[0036] To better understand the above-described objectives, features, and advantages of the embodiments of the present invention, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0037] Many specific details are set forth in the following description in order to provide a full understanding of this application. However, embodiments of the invention 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.

[0038] In actual navigation operations, a ship's planned transport route is usually from the port of origin directly to the port of destination. However, due to reasons such as refueling stops, crew replenishment, equipment maintenance, and waiting for navigation, the ship's actual navigation path is often naturally divided into multiple segments. Viewing only the trajectory of a single ship does not provide a clear, intuitive, and logical representation of the route from one port to another.

[0039] Traditional route calculation methods mainly rely on established nautical charts and historical navigation experience for path planning. While such methods can quickly obtain theoretical routes, from the perspective of digital applications, exhaustively searching for all global routes based on experience is not only extremely labor-intensive but also prone to errors, potentially leading to inaccuracies in voyage prediction, cost calculation, and timeliness assessment.

[0040] Furthermore, static path calculation methods cannot effectively handle time-series data in dynamic shipping environments and are unable to reflect the impact of real-world factors such as different months, different ship types (e.g., container ships, bulk carriers, tankers), and different ship tonnages.

[0041] Against this backdrop, by utilizing information such as latitude and longitude trajectories, port berthing events, and timestamps recorded by historical Automatic Identification Systems (AIS), combined with Marine Geographic Information Systems (MarineGIS) and route optimization algorithms, multiple voyage segments are identified, data are fused, and routes are reconstructed. The original discrete voyage segments are merged into continuous and complete cross-regional routes. Based on this, the actual sailing distance and route from the port of origin to the port of destination are calculated. This has significant value for improving the accuracy of voyage assessment, optimizing route design, and reducing transportation costs.

[0042] The relevant technologies have been applied to river shipping routes, big data analysis, and dynamic route planning, but they have not adopted the approach of combining port events with historical AIS data, and have failed to build a complete model for the fusion of multi-port data to solve the planning and design problems of long-distance, multi-port intercontinental routes.

[0043] The design and planning of shipping routes between any two ports globally are mostly based on static geographical path models, often employing a single origin and destination calculation method, lacking in-depth consideration of time cycles and individual differences in vessel operations. Existing methods do not refine route calculations and assessments according to different months, vessel types (such as container ships, bulk carriers, and tankers), vessel tonnage, and their technical characteristics. This results in generated route data and predictions that cannot be accurately adapted to specific vessel groups, limiting the application value of route planning results in diversified fleet management and refined industry operations.

[0044] Furthermore, the relevant research methods based on historical AIS data of ships mainly focus on historical port berthing information of ships and continuous AIS time series data in order to reconstruct the route between the two ports for analysis and planning.

[0045] However, in long-haul scenarios spanning intercontinental or cross-regional routes, this approach is largely limited to path analysis within a single segment or limited area, and a data fusion system for long-haul routes with multiple port stops has not yet been established. Existing technologies fail to effectively integrate port events (such as refueling, maintenance, and navigation waiting times) with historical AIS navigation trajectories, lacking fusion algorithm models for cross-port, multi-segment voyages, and are unable to construct complete operational paths and distance assessments for intercontinental routes on a global scale. This deficiency is particularly prominent in the scientific planning of ocean shipping routes, hub port connections, and complex shipping networks.

[0046] Based on the aforementioned background and related technical deficiencies, current route planning, especially the design and analysis of long-haul intercontinental routes, lacks dynamic and refined data-driven capabilities, making it difficult to accurately reflect different vessel attributes, port stop events, and long-distance, multi-segment operating routes.

[0047] Furthermore, existing research methods based on AIS data are generally limited to trajectory analysis of direct voyages between two ports, failing to form a fusion model of port event data and historical AIS trajectory data, thus leaving a significant gap in the comprehensive analysis of cross-port, multi-segment voyages.

[0048] This invention aims to realize a route reconstruction method with data fusion and dynamic attribute analysis as its core, which can enhance the accuracy of route calculation and improve the intelligence and scientific level of shipping management.

[0049] The following detailed description of the multi-port route reconstruction method, apparatus, electronic device, readable storage medium, and chip provided in this application embodiment, through specific embodiments and application scenarios, is provided in detail.

[0050] This embodiment provides a method for reconstructing shipping routes between multiple ports, such as... Figure 1 As shown, the route reconstruction method includes:

[0051] Step S100: Obtain the preset port sequence and port event database;

[0052] Step S102: Select a list of sample vessels that meet the port sequence within the set observation period from the port event database. The port sequence includes the port of origin, at least one intermediate port, and the port of destination.

[0053] Step S104: Based on the sample vessel list, extract the Automatic Identification System (AIS) data of at least one sample vessel from the port of origin to the port of destination;

[0054] Step S106: Process the data from the Automatic Identification System (AIS) to identify multiple navigation trajectory segments of the sample vessel between the port of origin, at least one intermediate port, and the port of destination.

[0055] Step S108: Merge and reconstruct multiple navigation trajectory segments to determine a continuous route from the port of origin to the port of destination;

[0056] Step S110: Determine the route reconstruction path from the origin port to the destination port based on multiple consecutive routes, and associate and store the route reconstruction path with the attribute information of at least one sample vessel.

[0057] Among them, multiple consecutive routes in the route reconstruction path pass through at least one different intermediate port.

[0058] The route reconstruction method provided by this invention selects eligible sample vessels and their historical Automatic Identification System (AIS) data from a port event database using a preset sequence including a port of origin, at least one intermediate port, and a port of destination. Then, by processing the AIS data, the method identifies the navigation trajectory segments of the sample vessels between multiple ports and merges these segments to reconstruct a continuous route from the origin to the destination. Finally, these continuous routes are associated with and stored with vessel attribute information, thereby constructing a dynamic and refined route reconstruction path. This reconstructed route includes multiple continuous routes passing through different intermediate ports, rather than a single static path.

[0059] The port event database includes sample ships corresponding to different months, different ship types (such as container ships, bulk carriers, tankers, etc.), and different ship tonnages.

[0060] For example, route data obtained through route reconstruction can not only provide accurate references for voyage assessment in chartering operations, but also be applied to the predictive analysis of Estimated Time of Arrival (ETA), maritime cost accounting, and route optimization decisions. Route data obtained through route reconstruction can shorten calculation time while ensuring calculation accuracy.

[0061] By integrating information on multiple voyage segments and stopover ports, continuous generation of intercontinental routes is achieved, enabling the acquisition of complete and accurate route and distance information from the port of origin to the port of destination even when vessels stop at multiple ports. The route reconstruction methods, which generate reconstructed routes including multiple continuous routes, provide more accurate voyage data support for ship leasing, port scheduling, route optimization, and other ship-related businesses, reducing costs and errors in long-haul planning.

[0062] For example, port events include, but are not limited to, ship refueling, maintenance, and navigation delays.

[0063] Understandably, this invention introduces joint analysis of port events and historical AIS data into the field of route design, and integrates multiple voyage segments through data fusion technology to construct a complete, continuous operating path across ports and multiple segments, thereby enabling the discovery, generation, and extraction of intercontinental routes. Even when vessels stop at multiple ports, the entire route is reconstructed and precise distance and path calculations are performed, overcoming the previous limitation of only being able to handle direct voyages between two ports. This enhances the accuracy of route calculations while improving the intelligence and scientific level of shipping management.

[0064] Specifically, during actual long-distance or intercontinental voyages, the ship's actual navigation trajectory is often naturally divided into multiple segments due to refueling stops, crew replenishment, equipment maintenance, and waiting for navigation. However, in the process of ship route planning, it is necessary to calculate the actual navigation distance and route from the port of origin to the port of destination. By segmenting and identifying multiple voyage segments, fusing data, and reconstructing routes, the originally discrete segments are merged into continuous and complete inter-regional routes, i.e., multiple continuous routes.

[0065] In historical port event databases for long-distance or intercontinental voyages, the number of route data for ships traveling directly from the port of origin to the port of destination is far less than the number of route data for ships departing from the port of origin and arriving at the port of destination via at least one intermediate port. Therefore, by utilizing the historical port event database for ships, a sample ship list can be formed by filtering ships that meet the port sequence of departing from the port of origin and arriving at the port of destination via at least one intermediate port.

[0066] It is understandable that the design and planning of shipping routes between any two ports are mostly based on static geographical path models, often employing a single method for calculating the origin and destination. However, in long-haul scenarios spanning intercontinental or cross-regional routes, this single method cannot effectively integrate port events with historical AIS navigation tracks, and cannot construct a complete operational path and distance assessment for intercontinental routes on a global scale. This paper addresses this issue by selecting a list of sample vessels that conform to port sequences within a set observation period from a port event database, thereby enabling the discovery, generation, and extraction of intercontinental shipping routes.

[0067] Furthermore, by accurately screening samples and extracting the entire trajectory based on preset port sequences, the amount of samples that can be processed and the number of routes that can be generated can be increased.

[0068] The sample vessel list is compiled by aggregating all vessels whose actual navigation trajectories conform to the aforementioned preset port sequence within a set observation period (e.g., three months). This improves the real-time performance of the route reconstruction paths obtained through AIS data reconstruction and fusion, meeting the actual vessel route planning needs.

[0069] Among these measures, the observation period is dynamically adjusted by relevant personnel in the ship route planning process based on actual planning needs.

[0070] After identifying multiple sample vessels that meet the preset port sequence based on the sample vessel list, the AIS data of multiple sample vessels during their respective entire voyages are scheduled to determine the arrival and departure timestamps of the sample vessels.

[0071] The AIS data of the sample vessels also includes detailed information such as the vessel's latitude, longitude, speed, and course throughout the entire voyage (from departure from the port of origin, through at least one intermediate port to the port of destination).

[0072] During the extraction of Automatic Identification System (AIS) data from at least one sample vessel from its port of origin to its port of destination, the extraction results are stored separately for each sample vessel to ensure that subsequent analysis can be performed in batches or as a single target based on individual vessel trajectories.

[0073] The Automatic Identification System (AIS) data of a single sample vessel is processed to divide the vessel's trajectory into multiple segments, resulting in multiple navigation trajectory segments between the port of origin, at least one intermediate port, and the port of destination. Based on preset speed thresholds, the AIS data of the sample vessel is analyzed to identify segments indicating low-speed drifting or port stops, thus obtaining multiple navigation trajectory segments.

[0074] Based on the speed threshold, the trajectory of a single ship is divided into a low-speed segment and a navigation segment, with the low-speed segment corresponding to the area where the ship stays in the port.

[0075] After identifying multiple segments, the segment information of sample vessels passing through the same intermediate port is integrated and analyzed to generate a continuous route from the port of origin to the port of destination. In the process of merging and reconstructing the navigation trajectory segments, the messy and curved trajectories of sample vessels when entering and leaving the port and the abnormal points in the intermediate segments are removed, and a clean and continuous route trajectory is retained.

[0076] Ultimately, a reconstructed route from the port of origin to the port of destination is determined through multiple consecutive routes, with each consecutive route passing through at least one different intermediate port. In other words, the reconstructed route from the port of origin to the port of destination includes multiple routes via different intermediate ports, enriching the diversity and practicality of vessel course planning.

[0077] Furthermore, each generated route is bound to the attribute information of the sample vessels that generated the continuous route. It can generate refined and differentiated route models based on attribute information such as different months, seasonal characteristics, vessel categories and tonnage, to meet the route analysis needs under different operating scenarios.

[0078] The system automatically builds differentiated route databases for different months, vessel types, and tonnage classes, making route plans more adaptable to actual operating conditions in terms of seasonal changes, vessel operating characteristics, and carrying capacity. For example, it provides sheltered routes for oil tankers and deep-water channel selection for large container ships, thereby improving operational safety and economy.

[0079] In some embodiments, the order in which at least one intermediate port appears in the port sequence is preset to be variable, and / or the intermediate port is selected from a predefined set of alternative ports to extend the route trajectory on multiple different headings during long-distance or transcontinental voyages, enabling the route reconstruction path to cover a variety of complex routes and improving the flexibility of route planning.

[0080] In some embodiments, optionally, a conforming port sequence refers to the port event records of sample vessels within a set observation period, which strictly and sequentially include all ports in the port sequence, and do not call at any other ports outside the ports specified in the port sequence. This ensures that the selected vessel trajectories are highly pure and avoids interference from irrelevant port events.

[0081] In some embodiments, optionally, route suggestions are provided for one or more target vessels planning to sail from the port of origin to the port of destination, based on previously associated and stored route reconstruction paths. The provided route suggestions are generated based on route reconstruction paths that match the attribute information of the target vessels. By utilizing a well-constructed, refined route database, personalized recommendations are provided for specific voyages of specific vessels, improving the practicality and foresight of the route reconstruction method.

[0082] In some embodiments, optionally, during the processing of Automatic Identification System (AIS) data, the AIS data also includes marine environmental data synchronized with the AIS data in time. Data processing includes correcting or labeling navigation trajectory segments based on the marine environmental data. The marine environmental data includes at least one of the following: meteorological data, ocean current data, and wave data.

[0083] In some embodiments, the route reconstruction path may optionally include not only spatial trajectory coordinates but also derived statistical information associated with the trajectory points. This derived statistical information includes at least one of the following: the historical average airspeed of the waypoints, traffic frequency, and traffic preference for different seasons. By adding derived statistical information, the information density and decision support capabilities of the output data are improved.

[0084] In some embodiments, optionally, such as Figure 2 As shown, the data from the Automatic Identification System (AIS) is processed to identify multiple navigation trajectory segments of the sample vessel between the port of origin, at least one intermediate port, and the port of destination, including:

[0085] Step S1060: Perform outlier detection and cleaning on the Automatic Identification System (AIS) data to determine the preprocessed data;

[0086] Step S1062: Based on the preprocessed data, construct the main track corridor connecting the port of origin and the port of destination;

[0087] Step S1064: For a single sample vessel, determine at least one trajectory point corresponding to the sample vessel;

[0088] Step S1066: Determine the trajectory of a single ship based on at least one trajectory point falling within the corresponding range of the main track corridor;

[0089] Step S1068: Obtain the first ship speed threshold;

[0090] Step S1070: Determine multiple navigation trajectory segments corresponding to a single ship's trajectory based on the first ship speed threshold.

[0091] In this embodiment, the AIS data of the sample vessels is first cleaned to ensure data quality. Then, based on the trajectory points of multiple sample vessels, spatial clustering is used to identify areas with high point density, forming a strip-shaped main track corridor. The main track corridor reflects the historical main co-current channels of the sample vessels on a specific route. For a single sample vessel, each trajectory point is traversed to check whether the trajectory point falls within the range of the main track corridor. If the trajectory point is not within the corridor, it is considered an anomaly and removed. Based on at least one trajectory point falling within the corresponding range of the main track corridor, the individual trajectory of the sample vessel is determined.

[0092] Finally, based on the first ship speed threshold, the purified single ship trajectory is intelligently segmented to accurately identify multiple navigation trajectory segments of the ship between different ports.

[0093] Understandably, by eliminating outliers (based on velocity thresholds and determination within the main track corridor), combining outlier elimination with spatial clustering, a realistic strip-shaped track corridor can be constructed, improving the accuracy of segmented trajectory boundary positioning.

[0094] Specifically, outlier detection and removal are performed on the extracted AIS data, primarily removing speed outliers, position outliers, and heading outliers. Speed ​​outliers include trajectory points within the reasonable speed range for the sample vessel's corresponding ship type or route interval, as well as trajectory points that suddenly appear inland or far from the route. Position outliers include drift points, isolated points, and out-of-group position information corresponding to the sample vessel. Heading outliers include azimuth data that is discontinuous or abruptly changes direction.

[0095] By performing outlier detection and cleaning on the data from the Automatic Identification System (AIS), the method of preprocessing the data is determined, making the cleaned data more consistent with the distribution characteristics of the navigation trajectory and effectively reducing noise interference during subsequent modeling.

[0096] The track points of multiple sample ships after cleaning are input into a spatial clustering algorithm to identify areas where the track point density is higher than the track point density threshold, so as to form the main track corridor from the port of origin to the port of destination.

[0097] After determining the main channel corridor, all track points of a single sample vessel are traversed. If a track point falls within the main channel corridor, it is considered a reasonable navigation point; if it falls outside the main channel corridor, it is considered an anomaly and is removed. This effectively filters out track deviations caused by vessels avoiding obstacles, temporary anchoring, weather detours, or engaging in activities unrelated to the current voyage (such as temporary stops at non-target ports).

[0098] After obtaining the purified single-ship trajectory, the ship's navigation status or port dwell status is identified based on a preset first ship speed threshold (usually a low value, such as 2 knots). Multiple navigation trajectory segments corresponding to the single-ship trajectory are identified through the navigation status.

[0099] After determining the multiple navigation trajectory segments corresponding to a single ship's trajectory, each trajectory segment is further transformed into a standardized navigation event. A navigation event is defined by at least the event type (sailing / berthing), starting position, ending position, average speed, and duration. Multiple navigation trajectory segments are transformed into a time-ordered sequence of navigation events to bind the multiple navigation trajectory segments to at least one intermediate port, thus determining the time sequence in which the sample ship sails to different intermediate ports.

[0100] In some embodiments, optionally, when a single vessel track point is identified as deviating from the main track corridor, marine environmental data for the corresponding time point is acquired. Based on a causal inference model, the correlation between the deviation event and marine environmental data (such as encountering severe weather) is analyzed, and an explanatory label is generated for the deviation point.

[0101] In some embodiments, outlier detection and cleaning may optionally include using a trained anomaly detection model (such as an isolated forest or an autoencoder) to score AIS data for anomalies in real time and remove trajectory points with anomaly scores higher than a preset anomaly threshold.

[0102] In some embodiments, it may be optionally determined whether the travel distance between consecutive trajectory points far exceeds the theoretical maximum distance that can be reached based on the reported speed and the reporting time difference; if so, it is determined to be an abnormal jump point.

[0103] In some embodiments, the corridor width of the main track corridor can be dynamically adjusted to meet the route planning requirements of different precision and improve the flexibility of route reconfiguration.

[0104] In some embodiments, optionally, based on the point clusters obtained by spatial clustering, a rotated bounding box algorithm or a centerline extraction algorithm is used to generate directional vectorized channel centerlines and width information, rather than just a set of points.

[0105] In some embodiments, optionally, determining a single ship trajectory based on at least one trajectory point falling within the corresponding range of the main track corridor includes: assigning a confidence weight to the trajectory point based on its distance to the centerline of the main track corridor. For multiple consecutive trajectory points falling outside the corridor but with relatively small deviations, they are not directly eliminated, but rather marked as low-confidence points and assigned lower weights or smoothed in subsequent segmentation and reconstruction steps.

[0106] In some embodiments, optionally, such as Figure 3 As shown, multiple navigation trajectory segments corresponding to a single ship's trajectory are determined based on a first ship speed threshold, including:

[0107] Step S10702: Obtain the velocity parameters corresponding to the trajectory points;

[0108] Step S10704: Determine at least one low-speed segment based on the speed parameters and the first ship speed threshold, wherein the speed parameters of the trajectory points in the low-speed segment are less than the first ship speed threshold;

[0109] Step S10706: Determine the starting and ending points of the single-ship trajectory based on the low-speed section;

[0110] Step S10708: Determine multiple navigation trajectory segments from the trajectory start point and trajectory end point based on trajectory points whose speed parameters are less than the first ship speed threshold.

[0111] In this embodiment, a preset first ship speed threshold is used as a judgment criterion to automatically identify the low-speed segment in the trajectory segment in order to accurately locate the trajectory start and trajectory end in the sample ship's voyage, and to divide the single ship trajectory into multiple navigation trajectory segments according to the first ship speed threshold.

[0112] Understandably, traditional methods may require manual marking of port locations on nautical charts for segmentation, which is inefficient and highly subjective. This invention achieves precise, automated, and batch segmentation through a defined speed threshold. This ensures that different people processing the same batch of data at different times can obtain completely consistent results, making it highly reproducible and meeting the requirements of industrial applications.

[0113] Specifically, when a ship is docked in a port, its speed is extremely low or even zero. This significant characteristic signal can be used as a trajectory segmentation feature to automatically cut a continuous single-ship trajectory into multiple navigation trajectory segments.

[0114] The system automatically identifies port events of sample vessels based on speed characteristics, identifies low-speed drifting or port stay intervals based on a first vessel speed threshold (e.g., < 2 knots), and divides the trajectory of a single vessel into multiple navigation trajectory segments.

[0115] The process involves segmenting the trajectory of a single ship according to a time series, clustering continuous low-speed points to form low-speed segments, identifying the first low-speed segment as the trajectory starting point, and identifying the last low-speed segment as the trajectory ending point.

[0116] For example, each trajectory point in the AIS data contains speed information, which is used to distinguish whether the ship is in a berthing / low-speed maneuvering state or a normal navigation state by using a first ship speed threshold. All trajectory points with speeds below the first ship speed threshold are marked and clustered into low-speed segments. Low-speed segments typically correspond to various speed reduction operations of the sample ship in port, such as berthing, loading and unloading operations, or waiting at anchor.

[0117] The identified trajectory start and end points are used as the boundaries of the entire voyage, and all trajectory points between the trajectory start and end points with speeds higher than a first ship speed threshold are identified as sea navigation points. These sea navigation points are naturally divided into multiple continuous navigation trajectory segments by the low-speed segments (i.e., the stopping segments in intermediate ports) located between them.

[0118] In some embodiments, the first ship speed threshold is optionally not a fixed value, but is dynamically determined based on preset rules of ship type, ship tonnage, or the sea area in which the sample ship operates. The definition of low speed differs for different types of sample ships (such as large oil tankers and maneuverable tugboats). The perceived speed while anchored also differs in open and congested sea areas. Dynamically setting the first ship speed threshold improves the accuracy of segmenting the trajectory of a single ship.

[0119] In some embodiments, optionally, after determining at least one low-speed segment, the duration of the low-speed segment is further determined, and the low-speed segment is confirmed as an effective port stop segment only if the duration exceeds a preset minimum berthing time threshold.

[0120] In some embodiments, optionally, when determining the trajectory start point and trajectory end point based on the low-speed segment, the identified low-speed segment is overlaid with a known port geofence database, and only when the central area of ​​the low-speed segment coincides with the port geofence is it confirmed as a valid trajectory start point and trajectory end point.

[0121] In some embodiments, optionally, such as Figure 4As shown, multiple navigation trajectory segments are merged and reconstructed to determine a continuous route from the port of origin to the port of destination. This includes performing the following operations on navigation trajectory segments between two consecutive ports:

[0122] Step S1080: Obtain the second ship speed threshold, which is greater than the first ship speed threshold;

[0123] Step S1082: Remove trajectory points in the navigation trajectory segment whose speed is lower than the second ship speed threshold;

[0124] Step S1084: For trajectory breakpoints caused by removal, obtain the adjacent trajectory point pairs of the trajectory breakpoints;

[0125] Step S1086: Obtain the distance parameters and azimuth change parameters of adjacent trajectory point pairs;

[0126] Step S1088: Obtain the first preset tolerance threshold and the second preset tolerance threshold;

[0127] Step S1090: When the distance parameter is less than or equal to the first preset tolerance threshold and the azimuth change parameter is less than or equal to the second preset tolerance threshold, connect the adjacent trajectory point pairs.

[0128] Step S1092: When the distance parameter is greater than the first preset tolerance threshold, or the azimuth change parameter is greater than the second preset tolerance threshold, then based on the main track corridor, path supplementation is performed between adjacent track point pairs to complete the connection.

[0129] In this embodiment, the navigation trajectory segments of sample ships are fused and reconstructed based on speed thresholds and geometric features. A second ship speed threshold, greater than the first, is used to actively eliminate low-speed navigation trajectories near ports. Then, the reasonableness of connection points before and after the eliminated trajectory breaks is determined: if the trajectories on both sides of the break are continuous in distance and direction, they are directly connected; if not, based on the main channel corridor, an algorithm is used within a local area to fill in the connection points before and after the eliminated trajectory breaks. This seamlessly reconstructs multiple navigation trajectory segments interrupted by ports into a continuous route from the origin port to the destination port.

[0130] The algorithms used to complete the connection between adjacent trajectory point pairs that are adjacent to the trajectory breakpoint include, but are not limited to, the A* algorithm. The A* algorithm, as a heuristic search algorithm, is used to find the lowest-cost path across multiple nodes on a graphical plane.

[0131] It is understandable that by eliminating outliers (speed threshold, corridor determination) and reconstructing local paths (A-Star algorithm to fill in points), this invention can accurately restore the true shape of long-haul intercontinental routes while maintaining spatial continuity, thereby achieving more accurate calculation of flight distance and azimuth.

[0132] In some embodiments, optionally, if the distance parameter between adjacent track point pairs is greater than the local range corresponding to the main track corridor, the route reconstruction is determined to be unsuccessful, and multiple navigation track segments corresponding to the sample ship are removed as invalid data to reduce the amount of data processing and improve the efficiency of route reconstruction.

[0133] Specifically, a second ship speed threshold is set that is significantly higher than a first ship speed threshold for identifying sample ships' stays in port. The first ship speed threshold is used to detect sample ships' stay operations in port, while the second ship speed threshold is designed to locate sample ships' maneuvers initiating entry and exit from port near the port.

[0134] Before fully anchoring, the sample vessels undergo deceleration, turning, queuing, and waiting, with their corresponding navigation trajectories appearing as highly curved and dense coils on the electronic chart. By setting a second vessel speed threshold, all complex aspects of the navigation trajectory related to near-port operations can be filtered out.

[0135] After removing the curved navigation tracks caused by entering ports or staying in sea areas, one or more track breaks will appear on the original continuous navigation track connecting the two intermediate ports. The start and end points of multiple track breaks are the starting points for the ship to enter stable navigation after leaving the intermediate port, and the ending points for the ship to begin to leave stable navigation before approaching the next intermediate port.

[0136] Obtain the distance parameters, azimuth change parameters, and corresponding preset tolerance thresholds of adjacent trajectory point pairs. Determine whether path point supplementation is needed by judging the distance tolerance and azimuth change tolerance.

[0137] When the distance parameter is less than or equal to the first preset tolerance threshold and the azimuth change parameter is less than or equal to the second preset tolerance threshold, that is, when the distance parameter is less than or equal to the preset tolerance threshold corresponding to the distance parameter and the azimuth change parameter is less than or equal to the preset tolerance threshold corresponding to the azimuth change parameter, it means that the difference between the distance and the heading of the trajectories on both sides of the breakpoint is within the tolerance threshold range, and the adjacent trajectory point pairs adjacent to the trajectory breakpoint can be directly connected to form a continuous route.

[0138] When the distance parameter is greater than the first preset tolerance threshold, or the azimuth change parameter is greater than the second preset tolerance threshold, it indicates that the difference between the distance and heading of the trajectories on both sides of the breakpoint exceeds the tolerance threshold range. The main track corridor is invoked, and the algorithm is used in the local area corresponding to the main track corridor to fill in the adjacent trajectory point pairs adjacent to the trajectory breakpoint.

[0139] The main track corridor is a consensus channel formed by all sample ships navigating in the region. Within the spatial constraints of the main track corridor, a path search algorithm is used to find the shortest continuous route between adjacent track point pairs that is completely located within the track corridor, thus ensuring the authenticity and rationality of the generated path.

[0140] In some embodiments, the second vessel speed threshold and the preset tolerance threshold are optionally dynamically adjusted based on the vessel type, tonnage, or historical sea state data of the sample vessel's operating area. For example, large bulk carriers and small container ships have different maneuverability and different navigation rules in narrow waterways and open seas. By dynamically adjusting the second vessel speed threshold and the preset tolerance threshold, the accuracy of route reconstruction in different scenarios is improved.

[0141] In some embodiments, optionally, when filling in points on a path, different sub-regions within the main track corridor are assigned weights, and the path search algorithm will prioritize paths with higher frequency and greater weight for point filling.

[0142] In some embodiments, optionally, during the path completion process, the algorithm simultaneously references real-time or historical marine environmental data (such as ocean currents and wind direction), and the continuous routes generated by the completion process prioritize favorable route segments with the current or wind while satisfying spatial constraints.

[0143] In some embodiments, optionally, such as Figure 5 As shown, a continuous shipping route from the port of origin to the port of destination is determined, including:

[0144] Step S1094: Obtain the time series parameters of the port sequence;

[0145] Step S1096: Determine the sailing sequence of at least one intermediate port based on the timing parameters;

[0146] Step S1098: Determine the continuous route from the port of origin to the port of destination according to the sailing sequence.

[0147] In this embodiment, the timing parameters refer to the precise time information associated with the initial port, at least one intermediate port, and the destination port in the port sequence. The timing parameters are derived from the berthing and departure timestamps of each sample vessel at each port, recorded in the port event database. Determining the timing parameters using the berthing and departure timestamps to judge the sailing order at at least one intermediate port improves the realism and objectivity of the route reconstruction.

[0148] The actual navigation sequence of the sample vessel, from its port of origin to its destination via at least one intermediate port, is determined according to the navigation sequence. Furthermore, the navigation trajectory segments between adjacent ports are accurately segmented from the complete navigation data to form a continuous route.

[0149] The navigation sequence also includes situations where the same sample vessel visits the same intermediate port multiple times between the port of origin and the port of destination, or forms port sequence intersections in different voyages, and multiple consecutive routes with different navigation sequences are formed based on the situation of multiple visits to the same intermediate port or the situation of port sequence intersections in different voyages.

[0150] In some embodiments, optionally, the frequency of different sailing sequences between the port of origin and the port of destination for all sample vessels is statistically analyzed. The sailing sequence with the highest frequency is identified as the typical sequence, and the continuous routes generated by it are marked as typical routes. Sailing sequences with extremely low frequency are identified as abnormal sequences, and the generated low-frequency routes are selectively filtered out of the port event database or marked separately.

[0151] In some embodiments, optionally, such as Figure 6 As shown, the route reconstruction path is associated with and stored with the attribute information of at least one sample vessel, including:

[0152] Step S1100: Obtain the timestamp from the Automatic Identification System (AIS) data;

[0153] Step S1102: Determine the actual sailing time of the sample vessel based on the timestamp;

[0154] Step S1104: Obtain at least one vessel registration information corresponding to the sample vessel list;

[0155] Step S1106: Determine the basic attributes of the sample vessel based on the vessel registration information and the port event database;

[0156] Step S1108: Determine the attribute information of each consecutive route in the route reconstruction path based on the basic attributes and actual flight time;

[0157] Step S1110: Associate and store the attribute information of the continuous routes with the corresponding ships.

[0158] In this embodiment, by extracting and integrating the timestamps from the Automatic Identification System (AIS) data and the static attributes from the ship registration information, attribute information containing the voyage time dimension and the ship's own characteristics is generated. The attribute information is then accurately associated with and stored with the reconstructed continuous routes, thereby upgrading a single geometric route into a digital route rich in semantics that supports multi-dimensional queries and refined analysis.

[0159] Understandably, static route calculation methods cannot effectively handle time-series data in dynamic shipping environments. They fail to reflect the impact of real-world factors such as different months, different ship types (e.g., container ships, bulk carriers, tankers), and different ship tonnages. Consequently, the generated route data and prediction results cannot be accurately adapted to specific ship groups, limiting the application value of route planning results in diversified fleet management and refined industry operations.

[0160] Compared to existing static route models, this embodiment can automatically build differentiated route databases for different months, ship types, and tonnage classes, making route plans more adaptable to actual operating conditions in terms of seasonal changes, ship type operating characteristics, and carrying capacity.

[0161] Furthermore, the attribute-driven differentiated route library enables this embodiment to support more detailed operational strategy formulation, such as shelter routes for oil tankers and deep-water channel selection for large container ships, thereby improving operational safety and economy.

[0162] Specifically, each trajectory point in the AIS data carries a precise timestamp. By analyzing the entire time interval from the departure of a sample vessel from its port of origin through at least one intermediate port to its destination port, the actual sailing time of this voyage can be determined.

[0163] More importantly, it allows for the extraction of finer-grained information from the time interval, such as the specific year and month in which the sample ship voyages occurred, in order to determine the season to which the time interval belongs.

[0164] The static attribute information of a ship includes its unique identifier, ship type, ship tonnage, and other technical parameters.

[0165] The types of ships include, but are not limited to: container ships, bulk carriers, oil tankers, and liquefied natural gas carriers.

[0166] Ship tonnage includes, but is not limited to, gross tonnage and deadweight tonnage, which represent the size of the ship.

[0167] Other technical parameters include, but are not limited to, ship length, beam, and draft. These other technical parameters may affect the sample vessel's choice of channel depth.

[0168] The dynamic time and static ship attributes are integrated to form the attribute information of a sample ship, and the attribute information is uniquely bound to the reconstructed continuous route.

[0169] Users can view the attribute information of each consecutive route in the route reconstruction path in the database to understand the attributes of the route reconstruction path, and to view the seasonal patterns and operational preferences of different ship types.

[0170] For example, in the chartering business, reconstructing routes based on shipping routes can provide more accurate estimates of voyage distance and fuel consumption for specific vessel types and seasons.

[0171] For example, by reconstructing routes based on routes, companies can analyze the performance of routes under different attribute combinations and select the optimal route for fleet operations.

[0172] In one specific embodiment, the route reconstruction method may optionally include:

[0173] Step 1: Target Vessel and Route Instance Screening. Utilize the historical port event database to screen vessels that meet specific route patterns. Within a set observation period (e.g., three months), select vessels that operate on the route and do not call at other ports between two intermediate ports. The port sequence pattern can be represented as A → B → C → D.

[0174] The screening results form a sample ship list, providing an index basis for subsequent data extraction.

[0175] Step 2: AIS track data extraction. Based on the sample vessel list obtained in Step 1, use the timestamps of vessel arrival and departure events to extract the complete historical AIS data for each vessel from the port of origin (Port A) to the port of destination (Port D).

[0176] The extracted results are stored as separate files for each ship, ensuring that subsequent analysis can be performed in batches or as a single target based on individual ship trajectories.

[0177] Step 3: Outlier Detection and Data Cleaning. Outlier detection and removal are performed on the extracted AIS dataset, mainly including:

[0178] Speed ​​anomalies: such as points exceeding the reasonable speed range for this ship type or route section;

[0179] Location outliers: drift points, isolated points, and outlier location information;

[0180] Anomalies in heading: Azimuth data that is discontinuous or abruptly changes from the preceding and following points.

[0181] The cleaned data better matches the actual flight trajectory distribution characteristics, effectively reducing noise interference during subsequent modeling.

[0182] Step 4: Construction of Major Track Corridors. The cleaned track points are input into a spatial clustering algorithm, such as Density-Based Spatial Clustering of Applications with Noise (DBSCAN) or Gaussian Mixture Model (GMM), to identify areas with high point density, forming strip-shaped major track corridors. These major track corridors reflect the historical main passageways of ships on specific routes and can be visualized in a Geographic Information System (GIS).

[0183] Step 5: Determining and Removing Anomalies from a Single Ship's Track Corridor. For each ship's track, iterate through each track point and check if it falls within the main track corridor range established in Step 4. If a point is not within the corridor, it is considered an anomaly and removed. These points are typically portions of the track where the ship has deviated from its course or temporarily docked at a non-target port.

[0184] Step Six: Track Segmentation and Port Event Identification. The trajectory of a single ship is segmented according to the time series, and low-speed drifting or port stay intervals are identified based on speed thresholds (e.g., < 2 knots), and marked as T1, T2, ..., Tn.

[0185] Clustering is performed on the low-speed points in segment T1 to extract the location core as the trajectory starting point (Port of Start, Ps);

[0186] Clustering is performed on the low-velocity points in segment Tn, and the location core is extracted as the trajectory endpoint (Port of End, Pe).

[0187] Step 7: Outlier Removal and Path Reconstruction in Intermediate Segments. For the intermediate segment from T2 to T(n-1), perform the following processing:

[0188] Points with a speed of less than 10 knots can be removed based on practical evaluation, eliminating track curvature caused by entering ports or staying in sea areas;

[0189] Perform a reasonableness assessment on the connection pair (Pn, Pn+K) before and after the deleted point (i.e., the removed trajectory breakpoint):

[0190] 1) If the distance is ≤ 3 nautical miles and the azimuth change is ≤ 10°, connect directly;

[0191] 2) If the above conditions are not met, call the corridor core points constructed in step four, and use the A* algorithm to fill in the trajectory points within a local range (≤ 10 nautical miles);

[0192] 3) If the distance to the replacement point is greater than 10 nautical miles, the reconstruction is deemed to have failed.

[0193] Step 8: Comprehensive Association Construction of Vessel Attributes and Reconstructed Routes. After completing the trajectory reconstruction in Step 7, to further enhance the data value and analyzability of the intercontinental route database, the reconstructed route data and vessel attribute information will be comprehensively associated. This includes the following steps:

[0194] Time-dimensional information extraction: The actual voyage time is extracted from the timestamps of the ship's historical AIS data extracted in step two, mapping the time period from departure to arrival to a specific calendar month. This time-dimensional parameter enables analysis of route change trends based on seasonality or month.

[0195] Ship attribute data integration: Extract basic ship attributes from ship registration information and port event databases, including but not limited to: ship type (such as container ship, bulk carrier, tanker, liquefied gas carrier, etc.) and gross tonnage. Standardize and classify different types and tonnage groups for easier data analysis and model invocation.

[0196] Attribute and route data association mapping: The above-mentioned ship attribute information is associated and bound with the unique identifier (such as ship number, track file number) of the reconstructed route data output in step seven.

[0197] Add attribute fields to the route database so that each reconstructed route not only contains spatial trajectory data, but also multi-dimensional attribute information such as sailing month, ship type, and tonnage.

[0198] Understandably, compared with related technologies, the route reconstruction method based on port events and historical AIS data proposed in this invention has achieved significant technological breakthroughs in route generation paradigm, dynamic adaptation capability, and cross-port multi-segment fusion, and has the following beneficial effects:

[0199] (1) Paradigm shift in route design methodology and data-driven modeling:

[0200] Breaking with traditional models: This invention abandons the single model of relying on human experience and static historical navigation knowledge for cross-regional route planning. Instead, it uses massive amounts of historical AIS data of ships as the core input source and employs algorithmic processing and model-based learning to generate and optimize routes.

[0201] Improved data scale and coverage: Through precise sample screening (based on port event sequence rules) and full trajectory extraction, this invention has achieved an order-of-magnitude improvement in the amount of sample volume that can be processed and the number of routes that can be generated compared with related technologies. It is no longer limited to a limited set of manually analyzable datasets and has significantly expanded the coverage and diversity of routes.

[0202] Enhanced decision support: The data-driven generation model ensures that the route analysis results can more comprehensively reflect historical operational patterns, thereby providing a solid basis for decision-making in cross-regional navigation route planning, safety assessment, and scheduling optimization.

[0203] (2) Dynamic route generation and refined attribute adaptation capabilities:

[0204] Introducing multi-dimensional dynamic factors: In the route generation process, this invention not only considers spatial geometric features (trajectory corridor), but also introduces the time dimension (month) and ship attributes (type, tonnage, structural parameters), realizing a route generation mechanism that integrates multiple factors;

[0205] More targeted route generation: Compared with the existing static route model, this invention can automatically build differentiated route libraries for different months, ship types and tonnage classes, so that the route plan is more adapted to actual operating conditions in terms of seasonal changes, ship type operating characteristics and carrying capacity.

[0206] Refined shipping planning: The attribute-driven differentiated route library enables the invention to support more detailed operational strategy formulation, such as shelter routes for oil tankers and deep-water channel selection for large container ships, thereby improving operational safety and economy.

[0207] (3) Capability to construct intercontinental routes integrating multiple port segments:

[0208] Seamless reconstruction of multiple voyage segments: In trajectory processing, this invention can not only identify direct voyages between two ports, but also merge multiple voyage segments that span multiple ports and include docking and drifting segments to reconstruct a continuous full-journey route.

[0209] Precise distance and path calculation: By eliminating outliers (speed threshold, corridor determination) and reconstructing local paths (A-star algorithm to fill in points), this invention can accurately restore the true shape of long-haul intercontinental routes while maintaining spatial continuity, thereby achieving more accurate distance and azimuth calculation.

[0210] Breaking through the limitations of related technologies: Related technologies can generally only perform route analysis in the case of no stops or single-segment data, while this invention can still construct a continuous full-course track even when the ship stops at multiple ports or even across seasons, providing a new solution for the study and prediction of complex routes.

[0211] Overall, this invention is not only innovative and operable in its technical implementation, but also directly solves problems in the background technology such as high data noise, trajectory deviation, port stop interference, trajectory breakage, and the inability of static route databases to meet dynamic operational needs. Ultimately, it provides a richer, more accurate, and more intelligent technical foundation for improving the safety, optimizing the economy, and planning of cross-regional navigation.

[0212] like Figure 7 As shown in the illustration, this application embodiment also provides a multi-port route reconstruction device 900, which includes: a data acquisition module 902 for acquiring a preset port sequence and a port event database; a sample identification module 904 for filtering a list of sample vessels that conform to the port sequence within a set observation period from the port event database, wherein the port sequence includes a port of origin, at least one intermediate port, and a port of destination; and a data extraction module 906 for extracting Automatic Identification System (AIS) data of at least one sample vessel from its port of origin to its port of destination based on the list of sample vessels. The trajectory recognition module 908 processes data from the Automatic Identification System (AIS) to identify multiple navigation trajectory segments of a sample vessel between its port of origin, at least one intermediate port, and its port of destination. The trajectory fusion module 910 merges and reconstructs these multiple navigation trajectory segments to determine a continuous route from the port of origin to the port of destination. The route reconstruction module 912 determines a route reconstruction path from the port of origin to the port of destination based on the multiple continuous routes and associates and stores the route reconstruction path with the attribute information of at least one sample vessel. The at least one intermediate port traversed by the multiple continuous routes in the route reconstruction path is different.

[0213] The route reconstruction method is implemented through a multi-port route reconstruction device 900. Using a preset sequence including the origin port, at least one intermediate port, and the destination port, sample vessels meeting certain criteria and their historical Automatic Identification System (AIS) data are selected from the port event database. Then, by processing the AIS data, the navigation trajectory segments of the sample vessels between multiple ports are identified, and these multiple trajectory segments are merged and reconstructed into a continuous route from origin to destination. Finally, multiple continuous routes are associated with and stored with vessel attribute information, thus constructing a dynamic and refined route reconstruction path. Data fusion technology integrates multiple voyage segments to construct a complete, multi-port, continuous operating path, thereby enabling the discovery, generation, and extraction of intercontinental routes. Even when a vessel stops at multiple ports, the entire route is reconstructed and accurate distance and path calculations are performed, overcoming the previous limitation of only being able to handle direct voyages between two ports. This enhances the accuracy of route calculations and improves the intelligence and scientific level of shipping management.

[0214] like Figure 8 As shown, this application embodiment also provides an electronic device 1000, including a processor 1110, a memory 1109, and a program or instructions stored in the memory 1109 and executable on the processor 1110. When the program or instructions are executed by the processor 1110, they implement the various processes of the above-described route reconstruction method embodiment and achieve the same technical effect. To avoid repetition, they will not be described again here.

[0215] Optionally, the processor 1110 is used to acquire a preset port sequence and port event database;

[0216] Optionally, the processor 1110 is also configured to filter a list of sample vessels that conform to a port sequence within a set observation period from a port event database, the port sequence including the port of origin, at least one intermediate port and the port of destination.

[0217] Optionally, the processor 1110 is also configured to extract Automatic Identification System (AIS) data of at least one sample vessel from its port of origin to its port of destination, based on a sample vessel list.

[0218] Optionally, the processor 1110 is also used to process the Automatic Identification System (AIS) data to identify multiple navigation trajectory segments of the sample vessel between the port of origin, at least one intermediate port, and the port of destination.

[0219] Optionally, the processor 1110 is also used to fuse and reconstruct multiple navigation trajectory segments to determine a continuous route from the port of origin to the port of destination.

[0220] Optionally, the processor 1110 is also configured to determine a route reconstruction path from the port of origin to the port of destination based on multiple consecutive routes, and to associate and store the route reconstruction path with the attribute information of at least one sample vessel.

[0221] The memory 1109 can be used to store software programs and various data. The memory 1109 may primarily include a first storage area for storing programs or instructions and a second storage area for storing data. The first storage area may store the operating system, application programs or instructions required for at least one function (such as sound playback, image playback, etc.). Furthermore, the memory 1109 may include volatile memory or non-volatile memory, or both. The non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDRSDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus RAM (DRRAM). The memory 1109 in this embodiment includes, but is not limited to, these and any other suitable types of memory.

[0222] This application also provides a readable storage medium storing a program or instructions. When the program or instructions are executed by a processor, they implement the various processes of the above-described route reconstruction method embodiments and achieve the same technical effects. To avoid repetition, these will not be described again here. Furthermore, the readable storage medium improves the data storage capacity and data processing speed of the route reconstruction method in this application.

[0223] A computer-readable storage medium can be a tangible device that holds and stores instructions for use by an instruction execution device. A computer-readable storage medium can be, but is not limited to, electronic storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination of the foregoing. A non-exhaustive list of more specific examples of computer-readable storage media includes: portable computer floppy disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable optical disc read-only memory (CD-ROM), digital universal disk (DVD), memory cards, floppy disks, encoding mechanical devices (e.g., punched cards or grooves with raised structures for recording instructions), and any suitable combination of the foregoing. The computer-readable storage medium used herein should not be construed as the transmission of signals themselves, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through waveguides or other transmission media, or electrical signals transmitted through wires.

[0224] The processor is the processor in the electronic device described in the above embodiments. The readable storage medium includes computer-readable storage media, such as computer read-only memory (ROM), random access memory (RAM), magnetic disk, or optical disk.

[0225] This application also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled. The processor is used to run programs or instructions to implement the various processes of the above-described route reconstruction method embodiments and achieve the same technical effect. To avoid repetition, it will not be described again here. In addition, the chip improves the data processing speed corresponding to the route reconstruction method in this application.

[0226] It should be understood that the chip mentioned in the embodiments of this application may also be referred to as a system-on-a-chip, system chip, chip system, or system-on-a-chip, etc.

[0227] In this invention, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance; the term "multiple" refers to two or more unless otherwise explicitly defined. The terms "install," "connect," "link," and "fix" should be interpreted broadly. For example, "connect" can be a fixed connection, a detachable connection, or an integral connection; "link" can be a direct connection or an indirect connection through an intermediate medium. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0228] In the description of this invention, it should be understood that the terms "upper," "lower," "left," "right," "front," "rear," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or unit referred to must have a specific orientation or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0229] In the description of this specification, the terms "one embodiment," "some embodiments," "specific embodiment," etc., refer to specific features, structures, materials, or characteristics described in connection with an embodiment or example that are included in at least one embodiment or example of the present invention. In this specification, 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.

[0230] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for reconstructing shipping routes between multiple ports, characterized in that, include: Obtain the preset port sequence and port event database; The port event database is used to select a list of sample vessels that conform to the port sequence within a set observation period. The port sequence includes the port of origin, at least one intermediate port, and the port of destination. Based on the sample vessel list, extract the Automatic Identification System (AIS) data of at least one sample vessel from the port of origin to the port of destination; The data from the Automatic Identification System (AIS) is processed to identify multiple navigation trajectory segments of the sample vessel between the port of origin, at least one of the intermediate ports, and the port of destination. By merging and reconstructing multiple navigation trajectory segments, a continuous route from the port of origin to the port of destination is determined; The route reconstruction path from the port of origin to the port of destination is determined based on multiple consecutive routes, and the route reconstruction path is associated with and stored with the attribute information of at least one of the sample vessels. Wherein, at least one of the intermediate ports passed through by multiple consecutive routes in the route reconstruction path is different; The process of processing the data from the Automatic Identification System (AIS) to identify multiple navigation trajectory segments of the sample vessel between the port of origin, at least one of the intermediate ports, and the port of destination includes: The data from the Automatic Identification System (AIS) is subjected to outlier detection and cleaning to determine the preprocessed data. Based on the preprocessed data, a main shipping corridor connecting the port of origin and the port of destination is constructed. For a single sample vessel, determine at least one trajectory point corresponding to the sample vessel; The trajectory of a single ship is determined based on at least one of the trajectory points falling within the range corresponding to the main track corridor; Obtain the first ship speed threshold; Multiple navigation trajectory segments corresponding to the single ship trajectory are determined based on the first ship speed threshold. The process of merging and reconstructing multiple navigation trajectory segments to determine a continuous route from the port of origin to the port of destination includes performing the following operations on navigation trajectory segments between two consecutive ports: Obtain a second ship speed threshold, which is greater than the first ship speed threshold; Remove trajectory points in the navigation trajectory segment whose speed is lower than the second ship speed threshold; For trajectory breakpoints caused by removal, obtain the adjacent trajectory point pairs of the trajectory breakpoints; Obtain the distance parameters and azimuth change parameters of the adjacent trajectory point pairs; Obtain the first preset tolerance threshold and the second preset tolerance threshold; When the distance parameter is less than or equal to the first preset tolerance threshold and the azimuth change parameter is less than or equal to the second preset tolerance threshold, the adjacent trajectory point pairs are connected. When the distance parameter is greater than the first preset tolerance threshold, or the azimuth change parameter is greater than the second preset tolerance threshold, then based on the main flight path corridor, path supplementation is performed between the adjacent trajectory point pairs to complete the connection. The order in which at least one intermediate port appears in the port sequence is preset and variable, and / or the intermediate port is selected from a predefined set of candidate ports.

2. The route reconstruction method according to claim 1, characterized in that, The step of determining multiple navigation trajectory segments corresponding to the single ship trajectory based on the first ship speed threshold includes: Obtain the velocity parameters corresponding to the trajectory points; At least one low-speed segment is determined based on the speed parameters and the first ship speed threshold, wherein the speed parameters of the trajectory points in the low-speed segment are less than the first ship speed threshold; The starting and ending points of the single-ship trajectory are determined based on the low-speed segment. Multiple navigation trajectory segments are determined from the trajectory start point and the trajectory end point based on the trajectory points whose speed parameters are less than the first ship speed threshold.

3. The route reconstruction method according to claim 1, characterized in that, Determining a continuous shipping route from the port of origin to the port of destination includes: Obtain the time-series parameters of the port sequence; The sailing sequence of at least one of the intermediate ports is determined based on the timing parameters; A continuous route from the port of origin to the port of destination is determined based on the sailing sequence.

4. The route reconstruction method according to any one of claims 1 to 3, characterized in that, The step of associating and storing the reconstructed route with the attribute information of at least one of the sample vessels includes: Obtain the timestamp from the Automatic Identification System (AIS) data; The actual sailing time of the sample vessel is determined based on the timestamp. Obtain at least one vessel registration information corresponding to the sample vessel list; The basic attributes of the sample vessels are determined based on the vessel registration information and the port event database. The attribute information of each consecutive route in the route reconstruction path is determined based on the basic attributes and the actual flight time. The continuous routes are associated with and stored with the attribute information of the corresponding ships.

5. A route reconfiguration device between multiple ports, characterized in that, include: A data acquisition module, which is used to acquire a preset port sequence and port event database; The sample identification module is used to filter out a list of sample vessels that conform to the port sequence within a set observation period from the port event database. The port sequence includes the port of origin, at least one intermediate port, and the port of destination. The data extraction module is used to extract Automatic Identification System (AIS) data of at least one sample vessel from the port of origin to the port of destination based on the sample vessel list. The trajectory recognition module is used to process the data of the Automatic Identification System (AIS) and identify multiple navigation trajectory segments of the sample vessel between the port of origin, at least one of the intermediate ports, and the port of destination. A trajectory fusion module is used to fuse and reconstruct multiple navigation trajectory segments to determine a continuous route from the port of origin to the port of destination. The route reconstruction module is used to determine the route reconstruction path from the port of origin to the port of destination based on multiple consecutive routes, and to associate and store the route reconstruction path with the attribute information of at least one of the sample vessels. Wherein, at least one of the intermediate ports passed through by multiple consecutive routes in the route reconstruction path is different; The trajectory recognition module is also used to detect and clean outliers in the data of the Automatic Identification System (AIS) to determine preprocessed data; based on the preprocessed data, to construct a main channel corridor connecting the port of origin and the port of destination; for a single sample vessel, to determine at least one trajectory point corresponding to the sample vessel; to determine the trajectory of a single vessel based on at least one trajectory point falling within the range corresponding to the main channel corridor; to obtain a first vessel speed threshold; and to determine multiple navigation trajectory segments corresponding to the trajectory of the single vessel based on the first vessel speed threshold. The trajectory recognition module is further configured to: acquire a second ship speed threshold, which is greater than the first ship speed threshold; remove trajectory points in the navigation trajectory segment whose speed is lower than the second ship speed threshold; for trajectory breaks caused by removal, acquire adjacent trajectory point pairs; acquire distance parameters and azimuth change parameters of the adjacent trajectory point pairs; acquire a first preset tolerance threshold and a second preset tolerance threshold; when the distance parameter is less than or equal to the first preset tolerance threshold and the azimuth change parameter is less than or equal to the second preset tolerance threshold, connect the adjacent trajectory point pairs; when the distance parameter is greater than the first preset tolerance threshold or the azimuth change parameter is greater than the second preset tolerance threshold, perform path completion between the adjacent trajectory point pairs based on the main navigation corridor to complete the connection; The order in which at least one intermediate port appears in the port sequence is preset and variable, and / or the intermediate port is selected from a predefined set of candidate ports.

6. An electronic device, characterized in that, It includes a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the multi-port route reconfiguration method as described in any one of claims 1 to 4.

7. 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 multi-port route reconfiguration method as described in any one of claims 1 to 4.

8. A chip, characterized in that, The chip includes a processor and a communication interface, the communication interface being coupled to the processor, the processor being used to run programs or instructions to implement the steps of the multi-port route reconfiguration method as described in any one of claims 1 to 4.

Images