Intelligent dispatching method and system for ocean-going ships based on AIS space-time trajectory data

Abstract

The application discloses a kind of ocean-going ship intelligent scheduling method and system based on AIS space-time trajectory data, method includes: obtaining the AIS space-time trajectory data of multiple ships in target sea area, constructs the space-time association structure of multiple ships in the dimension of heading, speed and arrival time, constructs ship group cooperative behavior mode based on space-time association structure;Identify the space-time region that ship presents convergence deceleration, stay or dense convergence in predetermined time window, and dynamically generate virtual anchor area according to the stability and persistence of group behavior in space-time region;Virtual anchor area is used as implicit navigation constraint condition, and combined with ship group cooperative behavior mode to form joint constraint, potential navigation path prediction is carried out to ship with AIS signal missing or abnormal;Based on predicted trajectory, the scheduling of ocean-going ship is adjusted, and scheduling decision scheme is obtained.The application reduces the congestion risk and waiting time, improves the efficiency of maritime traffic and the intelligent level of scheduling.

Description

Technical Field

[0001] This application relates to the field of ship traffic management and intelligent navigation technology, and in particular to a method and system for intelligent scheduling of ocean-going vessels based on AIS spatiotemporal trajectory data. Background Technology

[0002] With the increasing density of ocean-going vessel traffic, ships frequently exhibit convergent deceleration, lingering, or dense converging in specific sea areas (such as waiting areas outside ports or near narrow channels), forming de facto temporary anchorages or waiting areas. However, existing anchorages are typically pre-defined, fixed geographical areas that cannot dynamically adapt to the real-time behavior of the vessel group. This makes it difficult to accurately predict potential navigation paths of ships in the event of missing or abnormal AIS (Automatic Identification System) signals (such as signal interference, equipment malfunction, or manual shutdown), thus affecting the accuracy of ship scheduling decisions.

[0003] Existing AIS trajectory processing technologies primarily focus on single-ship trajectory clustering, anomaly detection, trajectory interpolation prediction (such as Kalman filtering and cubic spline interpolation), or simple speed threshold identification of mooring behavior. However, they lack in-depth analysis of the collaborative behavior of multiple vessels, especially failing to consider multi-dimensional correlation analysis of heading change trends, speed change trends, and arrival time offsets. Furthermore, existing vessel scheduling methods largely rely on fixed anchorages and historical routes, neglecting dynamically generated implicit waiting areas as navigation constraints. This leads to biases in congestion risk assessment and resource allocation in scheduling schemes, easily causing vessel congestion, prolonged waiting times, or potential collision risks.

[0004] Therefore, there is an urgent need for an intelligent method that can mine the collaborative behavior patterns of ship groups from AIS spatiotemporal trajectory data, dynamically generate virtual anchorage areas, and apply them as implicit constraints to predict signal missing paths, ultimately optimizing the scheduling of ocean-going vessels. Summary of the Invention

[0005] This application provides a method and system for intelligent scheduling of ocean-going vessels based on AIS spatiotemporal trajectory data, which reduces congestion risks and waiting times, and improves maritime traffic efficiency and scheduling intelligence.

[0006] This application provides the following solution:

[0007] According to the first aspect, a method for intelligent scheduling of ocean-going vessels based on AIS spatiotemporal trajectory data is provided. The method includes: acquiring AIS spatiotemporal trajectory data of multiple vessels in a target sea area; performing temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data to construct a spatiotemporal correlation structure of multiple vessels in the dimensions of heading, speed, and arrival time; constructing a vessel group collaborative behavior pattern based on the spatiotemporal correlation structure; under the constraints of the vessel group collaborative behavior pattern, identifying spatiotemporal regions where vessels exhibit convergent deceleration, lingering, or dense convergence within a predetermined time window, and dynamically generating virtual anchorage areas based on the stability and continuity of group behavior within the spatiotemporal regions; using the virtual anchorage areas as implicit navigation constraints and combining them with the vessel group collaborative behavior pattern to form joint constraints; predicting potential navigation paths for vessels with missing or abnormal AIS signals to obtain the predicted trajectories of the vessels; and adjusting the scheduling of ocean-going vessels based on the predicted trajectories to obtain a scheduling decision scheme.

[0008] According to one achievable method in this application embodiment, the step of performing temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data to construct a spatiotemporal correlation structure for multiple ships in the dimensions of heading, speed, and arrival time includes: determining an adaptive time reference based on the transmission frequency and time interval distribution of AIS spatiotemporal trajectory data of multiple ships in the target sea area, and mapping the AIS spatiotemporal trajectory data of each ship to a unified temporal representation under the time reference; under the unified temporal representation, constructing a spatiotemporal correlation structure with ships as nodes and spatial proximity and similarity of navigation status as edge weights, wherein the similarity of navigation status is jointly determined by the trend of heading change, the trend of speed change, and the arrival time offset.

[0009] According to one achievable method in the embodiments of this application, the construction of a ship group collaborative behavior pattern based on the spatiotemporal correlation structure includes: within a predetermined time window, calculating the correlation between multiple ships in terms of heading change trend, speed change trend, and arrival time offset according to the spatiotemporal correlation structure; identifying a set of ships exhibiting synchronous change characteristics within the time window based on the correlation; evaluating the stability of the synchronous change of the ship set within a continuous time window; and determining the set of ships that maintains stable synchronous change within the continuous time window as the ship group collaborative behavior pattern.

[0010] According to one achievable method in this application embodiment, identifying the spatiotemporal region where ships exhibit convergent deceleration, stagnation, or dense convergence within a predetermined time window under the constraints of the ship group's cooperative behavior pattern includes: within the predetermined time window, jointly analyzing the rate of change of speed and the magnitude of change of course of multiple ships participating in the same ship group's cooperative behavior pattern to identify behavioral segments in which multiple ships exhibit synchronous deceleration or synchronous course stabilization within similar time periods; based on the synchronous behavioral segments, determining the spatial overlap region of multiple ships within the corresponding time period; evaluating the persistence of the spatial overlap region within a continuous time window; and determining the spatial overlap region that remains persistent within the continuous time window as the spatiotemporal region exhibiting convergent deceleration, stagnation, or dense convergence.

[0011] According to one achievable method in this application embodiment, the dynamic generation of the virtual anchorage area based on the stability and persistence of the group behavior within the spatiotemporal region includes: within the spatiotemporal region, statistically analyzing the change in the number of ships participating in the ship group's collaborative behavior mode over time, as well as the fluctuation range of the ships in terms of course and speed; based on the change and fluctuation range of the number of ships over time, calculating a stability index reflecting the stability of the group behavior; analyzing the trend of the stability index within a continuous time window to determine the persistence of the group behavior; and dynamically determining the spatial range, duration, and effective time period of the virtual anchorage area based on the stability index and persistence results.

[0012] According to one achievable method in this application embodiment, the step of using the virtual anchorage area as an implicit navigation constraint and combining it with the ship group cooperative behavior pattern to form a joint constraint to predict the potential navigation path of ships with missing or abnormal AIS signals, and obtaining the predicted trajectory of the ship, includes: based on the ship group cooperative behavior pattern, determining the typical heading change trend and speed change range of the ships participating in the group cooperative behavior within the corresponding time window; using the virtual anchorage area as a constraint condition for the potential dwelling or convergence area of ​​ships with missing or abnormal AIS signals, limiting the reachable space of the ship within the predicted time range; under the joint constraint of the typical heading change trend, speed change range, and reachable space, inferring the potential navigation path of the ship with missing or abnormal AIS signals within the predicted time range; and determining the potential navigation path as the predicted trajectory of the ship.

[0013] According to one achievable method in the embodiments of this application, the adjustment of the scheduling of ocean-going vessels based on the predicted trajectory to obtain a scheduling decision scheme includes: based on the predicted trajectory, identifying potential spatiotemporal conflict relationships between multiple ocean-going vessels and the virtual anchorage area and other vessel predicted trajectories within the predicted time range; generating an assessment result reflecting changes in vessel congestion risk or waiting time under different scheduling schemes based on the potential spatiotemporal conflict relationships; and adjusting the vessel scheduling order, navigation plan, and resource allocation strategy based on the assessment result to generate a scheduling decision scheme.

[0014] According to the second aspect, a smart scheduling system for ocean-going vessels based on AIS spatiotemporal trajectory data is provided. The system includes: a trajectory data acquisition unit, configured to acquire AIS spatiotemporal trajectory data of multiple vessels in a target sea area, perform temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data, construct a spatiotemporal correlation structure of multiple vessels in the dimensions of heading, speed, and arrival time, and construct a vessel group collaborative behavior pattern based on the spatiotemporal correlation structure; a virtual anchorage area generation unit, configured to identify spatiotemporal areas where vessels exhibit convergent deceleration, lingering, or dense convergence within a predetermined time window under the constraints of the vessel group collaborative behavior pattern, and dynamically generate virtual anchorage areas based on the stability and continuity of group behavior within the spatiotemporal areas; a predicted trajectory generation unit, configured to use the virtual anchorage areas as implicit navigation constraints and combine them with the vessel group collaborative behavior pattern to form joint constraints, predict potential navigation paths for vessels with missing or abnormal AIS signals, and obtain the predicted trajectory of the vessels; and a scheduling decision scheme generation unit, configured to adjust the scheduling of ocean-going vessels based on the predicted trajectory to obtain a scheduling decision scheme.

[0015] According to a third aspect, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0016] According to a fourth aspect, an electronic device is provided, comprising: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any one of the first aspects.

[0017] According to the specific embodiments provided in this application, the following technical effects are disclosed:

[0018] This application constructs a spatiotemporal correlation structure of ships in terms of heading, speed, and arrival time by performing temporal alignment and spatial correlation analysis on the spatiotemporal trajectory data of multiple ships in a target sea area. Based on this structure, stable ship group cooperative behavior patterns are discovered. Under the constraints of these patterns, spatiotemporal areas where ships tend to decelerate, stop, or converge densely are accurately identified, and virtual anchorage areas are dynamically generated as implicit navigation constraints. Furthermore, the potential paths of ships with missing or abnormal AIS signals are predicted using the group cooperative pattern, resulting in more accurate predicted trajectories. Finally, the ocean-going vessel scheduling decision-making scheme is optimized based on the predicted trajectories. This method effectively solves the problems of large path prediction deviations and difficulty in identifying dynamic waiting areas caused by missing signals in traditional technologies. It significantly improves the accuracy and reliability of ship trajectory prediction, reduces congestion risks and waiting times, and enhances maritime traffic efficiency and the level of intelligent scheduling, thus possessing significant practical application value.

[0019] Of course, any product implementing this application does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a system architecture diagram applicable to the embodiments of this application;

[0022] Figure 2 A flowchart of the intelligent scheduling method for ocean-going vessels based on AIS spatiotemporal trajectory data provided in this application embodiment;

[0023] Figure 3 A structural block diagram of an intelligent scheduling system for ocean-going vessels based on AIS spatiotemporal trajectory data provided in this application embodiment;

[0024] Figure 4 A schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0025] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of this application are within the scope of protection of this application.

[0026] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “a,” “the,” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0027] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0028] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0029] To facilitate understanding of this application, the system architecture on which this application is based will be described first. Figure 1 An exemplary system architecture that can be applied to embodiments of this application is shown, such as Figure 1 As shown, the system architecture may include: user equipment and an intelligent scheduling system for ocean-going vessels based on AIS spatiotemporal trajectory data located on the server side.

[0030] Users can input AIS spatiotemporal trajectory data through their user equipment, which then sends it to the server-side intelligent dispatching system for ocean-going vessels based on the AIS spatiotemporal trajectory data. The intelligent dispatching system for ocean-going vessels based on AIS spatiotemporal trajectory data can use the method provided in this embodiment to perform intelligent dispatching and adjust the dispatching decision scheme. The server can then send the dispatching decision scheme to the user terminal.

[0031] User devices can include, but are not limited to, smart mobile terminals, wearable devices, and PCs (Personal Computers). Smart mobile devices can include mobile phones, tablets, laptops, PDAs (Personal Digital Assistants), and connected cars. Wearable devices can include smartwatches, smart glasses, virtual reality devices, augmented reality devices, mixed reality devices, and so on.

[0032] A smart dispatching system for ocean-going vessels based on AIS spatiotemporal trajectory data can be configured as a standalone server, a server cluster, or a cloud server. A cloud server, also known as a cloud computing server or cloud host, is a host product within the cloud computing service system, designed to address the management difficulties and weak service scalability inherent in traditional physical hosts and Virtual Private Servers (VPS) services. Besides... Figure 1 In addition to the architecture shown, the intelligent scheduling system for ocean-going vessels based on AIS spatiotemporal trajectory data can also be set up on a computer terminal with strong computing power.

[0033] It should be understood that Figure 1 The user equipment and the intelligent dispatching system for ocean-going vessels based on AIS spatiotemporal trajectory data shown in the example are merely illustrative. Depending on the implementation requirements, any number of user equipment and the intelligent dispatching system for ocean-going vessels based on AIS spatiotemporal trajectory data can be implemented.

[0034] Figure 2 A flowchart illustrating an intelligent scheduling method for ocean-going vessels based on AIS spatiotemporal trajectory data, provided as an embodiment of this application. Figure 2 As shown, the method may include the following steps:

[0035] Step 201: Obtain AIS spatiotemporal trajectory data of multiple ships in the target sea area, perform temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data, construct a spatiotemporal correlation structure of multiple ships in the dimensions of heading, speed and arrival time, and construct a ship group collaborative behavior mode based on the spatiotemporal correlation structure;

[0036] Step 202: Under the constraints of the ship group cooperative behavior pattern, identify the spatiotemporal regions in which ships exhibit convergent deceleration, stop or dense convergence within a predetermined time window, and dynamically generate virtual anchorage areas based on the stability and continuity of the group behavior within the spatiotemporal regions.

[0037] Step 203: Using the virtual anchorage area as an implicit navigation constraint, and combining it with the ship group cooperative behavior pattern to form a joint constraint, the potential navigation path of ships with missing or abnormal AIS signals is predicted to obtain the predicted trajectory of the ship.

[0038] Step 204: Adjust the scheduling of ocean-going vessels based on the predicted trajectory to obtain a scheduling decision scheme.

[0039] As can be seen from the above process, this application constructs a spatiotemporal correlation structure of ships in terms of heading, speed, and arrival time by performing temporal alignment and spatial correlation analysis on the spatiotemporal trajectory data of multiple ships in the target sea area, and thereby mines stable ship group cooperative behavior patterns. Under the constraints of this pattern, the spatiotemporal areas where ships tend to decelerate, stop, or converge densely are accurately identified, and virtual anchorage areas are dynamically generated as implicit navigation constraints. Then, combined with the group cooperative pattern, potential path prediction is performed for ships with missing or abnormal AIS signals to obtain more accurate predicted trajectories. Finally, the ocean-going vessel scheduling decision-making scheme is optimized based on the predicted trajectory. This method effectively solves the problems of large path prediction deviation and difficulty in identifying dynamic waiting areas caused by missing signals in traditional technologies, significantly improves the accuracy and reliability of ship trajectory prediction, reduces congestion risk and waiting time, and improves maritime traffic efficiency and scheduling intelligence, which has important practical application value.

[0040] The following describes in detail each step of the above process and the effects that can be further produced, with reference to the embodiments. First, with reference to the embodiments, step 201, namely, "acquiring AIS spatiotemporal trajectory data of multiple ships in the target sea area, performing temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data, constructing a spatiotemporal correlation structure of multiple ships in the dimensions of heading, speed and arrival time, and constructing a ship group collaborative behavior mode based on the spatiotemporal correlation structure", will be described in detail.

[0041] First, acquire the spatiotemporal trajectory data of multiple vessels' Automatic Identification System (AIS) data within the target sea area. This data typically includes each vessel's position, speed, heading, and corresponding timestamps at different times. Due to differences in the transmission frequency of AIS equipment and maritime communication conditions, the times at which each vessel reports data are not uniform, and the time intervals also differ. Therefore, it is difficult to perform synchronous comparisons between multiple vessels directly using the raw data.

[0042] Therefore, these spatiotemporal trajectory data undergo time-series alignment processing. Specifically, based on the overall distribution characteristics of the transmission frequency and time interval of all ship data within the target sea area, an adaptive unified time reference is determined. Then, the trajectory points of each ship are mapped onto a unified temporal grid under this time reference, thereby achieving alignment of all ship trajectories in the time dimension, which facilitates subsequent synchronous analysis.

[0043] After completing temporal alignment, the method further conducts spatial correlation analysis. Based on a unified temporal representation, a spatiotemporal correlation structure is constructed with each ship as a node. The edge weights of this structure are jointly determined by the spatial proximity between ships and the similarity of their navigation states. Among them, the similarity of navigation states comprehensively considers multiple dimensions such as the trend of course change, the trend of speed change, and the time offset of reaching the same position, thus comprehensively characterizing the degree of correlation of ships in their navigation behavior. This structure not only reflects the proximity of ships in physical space, but also captures their cooperative characteristics in motion patterns.

[0044] As an feasible approach, temporal alignment and spatial correlation analysis are performed on AIS spatiotemporal trajectory data to construct a spatiotemporal correlation structure for multiple ships in terms of heading, speed, and arrival time. This includes: determining an adaptive time reference based on the transmission frequency and time interval distribution of AIS spatiotemporal trajectory data of multiple ships in the target sea area, and mapping the AIS spatiotemporal trajectory data of each ship to a unified temporal representation under the time reference; and constructing a spatiotemporal correlation structure with ships as nodes and spatial proximity and similarity of navigation status as edge weights under the unified temporal representation, wherein the similarity of navigation status is jointly determined by the trend of heading change, the trend of speed change, and the arrival time offset.

[0045] Specifically, the method first statistically analyzes the data transmission frequency and time interval distribution of all vessels within the target sea area, such as calculating the average interval, median interval, or dominant interval pattern, and determines an adaptive time reference based on this. This time reference can be dynamically adjusted according to the actual data characteristics, rather than using a fixed uniform interval, thereby preserving the integrity of the original information to the greatest extent and avoiding errors caused by over-interpolation. After determining the time reference, the original trajectory points of each vessel are mapped to the corresponding unified temporal grid of the reference through interpolation or resampling, achieving alignment of all vessel trajectories in the time dimension and forming a unified temporal representation.

[0046] Building upon temporal alignment, the method further conducts spatial correlation analysis to construct a spatiotemporal correlation structure for multiple ships. This structure is represented as a graph, where each ship is considered a node, and the edge weights connecting nodes comprehensively reflect spatial proximity and similarity in navigation status. Spatial proximity is primarily determined based on the geographical distance between ship positions; for example, if the distance between two ships at the same time point is less than a certain threshold, they are considered to be spatially correlated. Similarity in navigation status is a deeper characteristic, jointly determined by three aspects: course change trend, speed change trend, and arrival time offset. The course change trend reflects the consistency of ship turning, the speed change trend reflects the synchronicity of acceleration or deceleration, and the arrival time offset captures the probability that ships tend to arrive at the same area within a similar timeframe. This multi-dimensional similarity assessment can effectively identify ship pairs or groups that are not only physically close but also highly coordinated in their motion behavior.

[0047] The spatiotemporal correlation structure constructed through the above steps not only reflects the distribution relationship of ships in physical space, but more importantly, it incorporates the dynamic navigation characteristics after time synchronization, thus providing a structured and information-rich representation for subsequent exploration of ship group collaborative behavior patterns.

[0048] Finally, based on the spatiotemporal correlation structure constructed above, the method further mines and constructs ship group cooperative behavior patterns. By analyzing the highly correlated connections between nodes in the structure, it identifies ship sets that exhibit synchronized course adjustments, synchronized speed changes, and similar arrival time offsets within a certain time window. Through stability verification over continuous time windows, it identifies ship groups that consistently exhibit cooperative characteristics. This group cooperative behavior pattern reflects phenomena commonly seen in actual maritime traffic, such as convoy following, collective waiting, or joint avoidance, providing a reliable basis for subsequent identification of hidden waiting areas.

[0049] As an implementable approach, constructing a ship group collaborative behavior pattern based on the aforementioned spatiotemporal correlation structure includes: within a predetermined time window, calculating the correlation between multiple ships in terms of heading change trends, speed change trends, and arrival time offsets according to the aforementioned spatiotemporal correlation structure; identifying a set of ships exhibiting synchronous change characteristics within the time window based on the correlation; evaluating the stability of the synchronous change of the ship set within a continuous time window; and determining the ship set that maintains stable synchronous change within the continuous time window as the ship group collaborative behavior pattern.

[0050] Specifically, within a predetermined time window, the correlation between multiple ships is calculated based on their spatiotemporal relationship structure. Specifically, correlation coefficients or other similarity indicators are calculated for three dimensions: course change trend, speed change trend, and arrival time offset. The course change trend reflects whether ships tend to turn in unison, the speed change trend reflects whether ships accelerate or decelerate simultaneously, and the arrival time offset reflects whether ships converge towards the same area within a similar timeframe. This multi-dimensional correlation calculation comprehensively characterizes the synchronicity of ships' motion patterns, avoiding the limitations of single-dimensional judgments, and thus more accurately identifying potential cooperative relationships.

[0051] Next, based on the calculated correlation, sets of vessels exhibiting synchronous changes within the current time window are identified. Typically, by setting a correlation threshold or using clustering algorithms, vessels with high correlation are grouped into the same set. These sets represent groups of vessels whose navigational behavior is highly coordinated within that time window, such as convoys collectively slowing down to wait for port entry, groups of vessels coordinating avoidance, or groups of fishing boats collectively loitering. This identification process transforms scattered individual correlations into clear group units, providing objects for subsequent stability assessments.

[0052] The method then assesses the stability of synchronous changes within consecutive time windows for the initially identified vessel assemblies. Due to the complexity of the maritime traffic environment, brief synchronous changes may be merely accidental phenomena, such as speed adjustments during brief vessel crossings, while genuine group cooperative behavior should exhibit a certain degree of continuity. Therefore, by using multiple sliding or consecutive time windows, the method continuously monitors the fluctuations in the correlation of the vessel assemblies across three dimensions, calculating its stability indices, such as the variance, standard deviation, or rate of change of the correlation coefficient. Only vessel assemblies that maintain high correlation and low fluctuations across multiple consecutive time windows are considered to possess stable synchronous change characteristics.

[0053] Finally, the vessel ensemble that passed the stability assessment was formally identified as the vessel group cooperative behavior pattern. This pattern reflects persistent collective behaviors commonly seen in actual navigation, such as convoy following, collective waiting outside ports, or group loitering in specific operational areas. It not only captures the instantaneous connections between vessels but also emphasizes the continuity and reliability of behavior, providing a solid basis for subsequently identifying spatiotemporal areas of convergent deceleration, lingering, or dense convergence under the constraints of this pattern.

[0054] The following describes in detail step 202, namely, "under the constraints of the ship group's collaborative behavior pattern, identifying the spatiotemporal regions in which ships exhibit convergent deceleration, lingering, or dense convergence within a predetermined time window, and dynamically generating a virtual anchorage region based on the stability and continuity of the group behavior within the spatiotemporal region," with reference to the embodiments.

[0055] First, using a ship group cooperative behavior pattern as a constraint, behavioral analysis is conducted on multiple ships participating in this pattern within a predetermined time window. The focus is on whether these ships collectively exhibit convergent deceleration, stagnation, or dense convergence within similar timeframes. Convergent deceleration is characterized by multiple ships reducing their speed almost simultaneously; stagnation is characterized by ships maintaining a low speed within a limited area for an extended period; and dense convergence is characterized by a significant increase in ship density, forming a de facto temporary cluster. This identification process is not simply based on the speed threshold of a single ship, but is strictly limited to ship groups with confirmed stable cooperative relationships, thereby significantly reducing misjudgments of passing ships or occasional deceleration behavior and improving the targeting and accuracy of the identification.

[0056] The identification process involves a joint analysis of the rate of change of speed and the magnitude of change of course of ships within the group. By simultaneously detecting whether multiple ships exhibit synchronous deceleration or synchronous course stabilization, corresponding spatial overlap areas are identified. Subsequently, the persistence of these spatial overlap areas within consecutive time windows is evaluated. Only those areas that maintain a high degree of overlap and group behavioral characteristics across multiple consecutive time windows are ultimately identified as exhibiting convergent deceleration, stagnation, or dense convergence. This persistence assessment ensures the reliability of the identification results, avoiding the misidentification of transient traffic phenomena as stable waiting or anchoring behavior.

[0057] As an implementable approach, this application identifies spatiotemporal regions where ships exhibit convergent deceleration, stagnation, or dense convergence within a predetermined time window under the constraints of the ship group's cooperative behavior pattern. This includes: within the predetermined time window, jointly analyzing the rate of change of speed and the magnitude of change of course of multiple ships participating in the same ship group's cooperative behavior pattern to identify behavioral segments in which multiple ships exhibit synchronous deceleration or synchronous course stabilization within similar time periods; based on the synchronous behavioral segments, determining the spatial overlap region of multiple ships within the corresponding time period; evaluating the persistence of the spatial overlap region within a continuous time window; and identifying the spatial overlap region that remains persistent within the continuous time window as the spatiotemporal region exhibiting convergent deceleration, stagnation, or dense convergence.

[0058] Specifically, within a predetermined time window, a joint analysis of the rate of change of speed and the magnitude of change of course is conducted on multiple vessels participating in the same coordinated behavior pattern within a vessel group. The rate of change of speed is used to capture the dynamics of vessel acceleration or deceleration, while the magnitude of change of course reflects the severity of vessel turning. By jointly examining these two indicators, it is possible to identify behavioral segments in which multiple vessels simultaneously exhibit synchronous deceleration or synchronous course stabilization within a similar timeframe. Synchronous deceleration is characterized by a significant reduction in speed by most vessels within the group at similar times, while synchronous course stabilization is characterized by a small magnitude of change in course, typically indicating that the vessel has ceased to move in a straight line along its original direction and has entered a state of low-speed lingering or waiting in place. For synchronous deceleration, a rate of change of speed is typically set below a certain negative threshold, for example... Knots per minute (kJ / min) indicates significant deceleration. For synchronized heading stability, the heading change is typically set to be below a small threshold, for example... Degrees per minute (dF / min) indicates that the vessel hardly turns or only makes minor attitude adjustments. These thresholds can be determined based on statistical data from actual sea areas or experience, and can be adaptively adjusted during implementation. This joint analysis fully utilizes the constraints of group cooperative patterns, focusing only on vessels with confirmed stable cooperative relationships, thereby significantly improving the targeting and reliability of behavioral segment identification.

[0059] Next, based on the identified synchronous behavior segments, the method determines the spatial overlap area of ​​multiple vessels within the corresponding time period. Specifically, it performs overlay analysis on the position trajectories of all participating vessels within the behavior segment, calculating their overlapping portion in geographic space. The overlap boundary is typically delineated using methods such as convex hull algorithms, density clustering, or grid statistics, forming a clear spatiotemporal region. This region often corresponds to temporary waiting areas or de facto anchorage areas in actual maritime traffic, reflecting the dense spatial distribution of the vessel group due to deceleration or lingering.

[0060] The method then assesses the persistence of the aforementioned spatially overlapping regions within consecutive time windows. Due to the variability of the maritime environment, brief spatial overlaps may only be instantaneous phenomena during ship encounters, while genuine convergence deceleration, stopping, or dense convergence should have a longer duration. Therefore, by sliding through multiple consecutive time windows, indicators such as changes in the area of ​​the overlapping region, fluctuations in ship density, or shifts in center position are monitored. Only those overlapping regions that remain relatively stable across multiple windows and where ships continuously remain are considered to have sufficient persistence.

[0061] After identifying the aforementioned spatiotemporal regions, virtual anchorage areas are dynamically generated based on the stability and continuity of group behavior within these regions. Unlike traditionally preset fixed anchorages, virtual anchorage areas are generated entirely based on real-time vessel group behavior data, and their spatial range, duration, and effective time period can be dynamically adjusted according to actual conditions. Specifically, by statistically analyzing the changes in the number of vessels participating in the collaborative mode within the region over time, as well as the fluctuations in vessel heading and speed, a group behavior stability index is calculated, and the trend of this index within a continuous time window is analyzed to comprehensively determine the continuity of the behavior. Based on these quantitative results, the system dynamically delineates the boundaries and validity period of the virtual anchorage area, accurately reflecting the implicit waiting or dwelling characteristics of current maritime traffic.

[0062] As an implementable approach, the dynamic generation of a virtual anchorage area based on the stability and persistence of group behavior within the spatiotemporal region includes: within the spatiotemporal region, statistically analyzing the changes in the number of ships participating in the collaborative behavior pattern of the ship group over time, as well as the fluctuations in the ships' heading and speed; calculating a stability index reflecting the stability of group behavior based on the changes and fluctuations in the number of ships over time; analyzing the trend of the stability index within a continuous time window to determine the persistence of the group behavior; and dynamically determining the spatial range, duration, and effective time period of the virtual anchorage area based on the stability index and persistence results.

[0063] Specifically, within a defined spatiotemporal region, statistical analysis is conducted on vessels participating in the coordinated behavior patterns of the vessel group. This includes two aspects: first, the change in the number of vessels over time, such as fluctuations in the number of vessels entering, leaving, or remaining in the region within a continuous time window; and second, the fluctuation range of vessels in terms of course and speed, such as the standard deviation of course, the root mean square deviation of speed, or the difference between maximum and minimum speeds. These statistical indicators directly reflect the degree of aggregation and consistency of movement of the group within the region. Regions with stable and small fluctuations in the number of vessels usually indicate that the group is collectively waiting, while small fluctuations in course and speed further confirm that the vessels are in a low-dynamic state of inactivity.

[0064] Next, based on the aforementioned changes and fluctuations in the number of vessels, a stability index reflecting the stability of the group's behavior is calculated. This index can be a comprehensive quantitative value, such as a weighted sum after normalization, or a single score obtained using statistical methods such as entropy or coefficient of variation. A higher stability index indicates more orderly and consistent behavior of the group within the current spatiotemporal region, making it more suitable to be considered a stable temporary anchorage or waiting area. This calculation process transforms multi-source statistical data into a unified and comparable metric, providing an objective basis for subsequent sustainability assessments.

[0065] The method then analyzes the trend of stability indices over consecutive time windows to determine the persistence of group behavior. Specifically, by observing the numerical curves, slopes, or variances of stability indices across multiple adjacent time windows, it determines whether the indices remain at a high level with minimal fluctuations. If the indices consistently exceed a preset threshold and exhibit a stable trend over a longer period, it indicates that the group behavior is sufficiently persistent; conversely, if the indices decline rapidly or fluctuate drastically, it suggests that the aggregation phenomenon in the area is merely a transient traffic event. This trend analysis ensures that virtual anchorage areas are not incorrectly generated due to transient phenomena, thereby improving the reliability of the generation process.

[0066] Finally, based on stability indicators and persistence analysis results, the spatial extent, duration, and effective period of the virtual anchorage area are dynamically determined. The spatial extent can be defined by the convex hull, polygonal boundaries, or density contour lines of the current vessel position distribution, and is adjusted in real time as the number and position of vessels change. The duration and effective period are based on persistence trend predictions, for example, from when behavior begins to stabilize until the indicators begin to decline significantly. This dynamic generation mechanism allows the virtual anchorage area to fully adapt to real-time traffic conditions, offering greater flexibility and accuracy compared to traditional fixed anchorages.

[0067] Overall, this technology, through precise identification and dynamic generation mechanisms under the constraints of a group collaborative mode, effectively captures temporary waiting areas outside traditional fixed anchorages. The introduction of virtual anchorage areas transforms the actual behavior of ship groups into quantifiable navigation constraints, not only compensating for the neglect of dynamic traffic phenomena in existing technologies but also providing more realistic implicit evidence for path prediction in the absence of AIS signals, significantly improving the practicality and intelligence level of the entire intelligent scheduling method.

[0068] The following describes in detail step 203, namely, "using the virtual anchorage area as an implicit navigation constraint and combining it with the ship group's collaborative behavior pattern to form a joint constraint, predicting the potential navigation path of ships with missing or abnormal AIS signals, and obtaining the predicted trajectory of the ship," with reference to the embodiments.

[0069] First, the method uses a virtual anchorage area as an implicit navigation constraint. The virtual anchorage area represents the temporary stopping or waiting area actually formed by the vessel group. For vessels with missing or abnormal signals, even if their own tracks are interrupted, this area can still be considered a potential stopping or convergence target. This means that when predicting paths, the reachable space of vessels within the predicted time frame must be limited, causing the predicted trajectory to tend to enter or stay within this area for a period of time, rather than arbitrarily deviating from group behavior. This implicit constraint avoids directly assuming that vessels have a specific destination, but rather provides a more natural navigation constraint based on the collective factual behavior of surrounding vessels.

[0070] Simultaneously, the method incorporates a joint constraint based on the collaborative behavior pattern of the vessel group. Specifically, it extracts typical course change trends and speed change ranges of participating vessels within corresponding time windows from this pattern. These trends and ranges reflect the overall movement patterns of the group; for example, most vessels are slowly adjusting their course or lingering in low-speed ranges. For vessels lacking signals, since they are likely to belong to the same group or be affected by the same traffic, their potential paths should also follow similar course trends and speed ranges. This constraint ensures that the predicted trajectory is not only spatially reasonable but also dynamically consistent with surrounding vessels.

[0071] Under the aforementioned joint constraints, the method infers potential navigation paths for ships with missing or abnormal signals. Algorithms such as particle filtering, Bayesian inference, or constraint optimization are typically employed to limit the reachable space while forcing the predicted path's heading and speed sequences to conform to typical population trends and ranges. Through multiple sampling or iterative optimizations, several candidate paths that meet the constraints are generated. Finally, one or more paths with the highest probability or best matching the population characteristics are selected as potential navigation paths and determined as the ship's predicted trajectory.

[0072] As an feasible approach, the virtual anchorage area is used as an implicit navigation constraint, and combined with the ship group cooperative behavior pattern to form a joint constraint. This allows for the prediction of potential navigation paths for ships with missing or abnormal AIS signals. The predicted trajectory of the ship includes: determining the typical heading change trend and speed change range of the ships participating in the group cooperative behavior within the corresponding time window based on the ship group cooperative behavior pattern; using the virtual anchorage area as a constraint on the potential stopping or convergence area of ​​ships with missing or abnormal AIS signals, limiting the reachable space of the ship within the predicted time range; inferring the potential navigation path of the ship with missing or abnormal AIS signals within the predicted time range under the joint constraints of the typical heading change trend, speed change range, and reachable space; and determining the potential navigation path as the predicted trajectory of the ship.

[0073] For typical course change trends and speed change ranges, the first step is to extract the course and speed sequences of participating vessels within the corresponding time windows. For a defined vessel group cooperative behavior pattern, one or more consecutive predetermined time windows (e.g., 10 minutes or 15 minutes) corresponding to that pattern are selected. For each vessel in the group... Extract its heading sequence within that time window. and speed sequence ,in To ensure consistent time points after time alignment. These sequences have undergone prior time alignment processing to ensure comparability of all ships on the same time grid.

[0074] The second step is to calculate the course change trend for each vessel. To calculate the course change trend, the course change rate sequence for each individual vessel within the window is typically calculated first. (To handle recurring problems, the minimum angle difference can be used), or the slope of the heading change over time can be directly fitted. Then, statistics are taken for the heading change rate sequence or fitted slope of all ships in the group. Common methods include: calculating the median or weighted average of the heading change rate of all ships at each time point within the window as the typical heading change trend of the group; or using the quantile method to obtain the confidence interval of the trend, such as the 25% to 75% quantile interval, to describe the typical magnitude and directionality of the group's heading adjustments. If the group as a whole tends to be stable, the typical trend is close to zero; if there is a consistent turning, it is represented by a positive or negative average rate of change.

[0075] The third step is to determine the speed variation range. For each speed variation range, first calculate the statistical characteristics of each vessel's speed within the window, such as average speed, standard deviation of speed, minimum / maximum speed, or rate of change of speed. Then, these characteristics of all vessels within the group are summarized: the typical speed variation range can be defined as the overall quantile range of all vessels' speeds within the window, such as the 10th to 90th percentile, or the mean ± 1.5 standard deviations, to exclude the influence of extreme values; a more precise approach is to statistically analyze the cross-distribution of speeds at each time point within the window, taking the kernel density estimate of the peak range of the group speed. This range reflects the range of low-speed lingering or slow adjustments maintained by the group during coordinated behavior, such as 2 to 6 knots.

[0076] The fourth step is to output typical features and apply them to subsequent constraints. The calculated typical heading change trends (e.g., average rate of change ± standard deviation) and speed change ranges (e.g., lower to upper limits) are used as dynamic feature descriptions of the group's cooperative behavior patterns. In a real system, these can be stored as parameter pairs for subsequent path prediction constraints on ships with missing signals, ensuring that the heading adjustment magnitude and speed values ​​of the predicted trajectory fall within this typical range.

[0077] The virtual anchorage area serves as a constraint on the potential dwelling or convergence area for vessels with missing or abnormal signals. The core of this constraint lies in limiting the reachable space of vessels within the predicted time frame. Specifically, even if a vessel's signal is interrupted, its predicted path is still required to enter or remain within the virtual anchorage area, or at least the reachable range of its path within the predicted time frame must not exceed the reasonable extension boundary of that area. This spatial limitation is based on the actual behavior of the group: since surrounding vessels have already established stable dwellings or waiting areas in this region, vessels with missing signals are highly likely to belong to the same group or be affected by the same traffic control measures, thus tending to slow down, linger, or remain in the same area. This implicit constraint avoids the problem of arbitrary path diffusion after signal interruption in traditional prediction methods, making the prediction results more consistent with actual maritime traffic patterns.

[0078] Secondly, under the joint constraints of typical course change trends, speed change ranges, and the aforementioned reachable space, the method infers the potential navigation paths of vessels with missing signals or abnormalities within the prediction time range. Typical course change trends and speed change ranges originate from group cooperative behavior patterns, representing the motion characteristics of most vessels within the group during the corresponding time period, such as small course adjustments and speeds maintained in the low-speed range. During prediction, the system requires the generated path sequence to simultaneously meet three conditions: the rate of change of course conforms to the typical trend range, the speed value falls within the typical range, and the path as a whole lies within the reachable space defined by the virtual anchorage area. Multiple candidate paths are typically generated using constraint optimization algorithms, particle filtering, or Monte Carlo sampling techniques, and then the best-fitting path sequence is obtained through screening or weighted averaging. This joint constraint mechanism fully utilizes collective intelligence, ensuring that the predicted path is not only spatially reasonable but also exhibits dynamic behavior highly consistent with surrounding vessels.

[0079] Finally, the inferred potential navigation path is determined as the predicted trajectory of the vessel. This predicted trajectory can be a single optimal path or a set of multiple paths with probability distributions, depending on the application scenario. It directly serves subsequent vessel scheduling decisions, more accurately reflecting the true movement intentions and future positions of vessels lacking signals.

[0080] Overall, this technology cleverly combines the spatial constraints of the virtual anchorage area with the dynamic constraints of the group's collaborative behavior pattern, enabling intelligent trajectory completion for vessels lacking signals. It fully utilizes the collective information of surrounding vessels as "collective intelligence," making the prediction results closer to the actual navigation intentions and avoiding the significant deviations that may arise from relying solely on kinematic models or historical trajectories. This joint constraint mechanism provides high-quality input for subsequent vessel scheduling adjustments based on predicted trajectories, demonstrating the method's advanced nature and practical value in handling complex maritime traffic scenarios.

[0081] The following describes step 204, namely "adjusting the scheduling of ocean-going vessels based on the predicted trajectory to obtain a scheduling decision scheme," in detail with reference to the embodiments.

[0082] First, based on the predicted trajectories of all vessels, including the actual trajectories of vessels with normal signals and the predicted trajectories of vessels without signals, a comprehensive simulation of the future motion of multiple ocean-going vessels within the target sea area is conducted. These predicted trajectories allow for the early identification of potential spatiotemporal conflicts that may arise between vessels within the predicted timeframe. For example, if the predicted paths of two vessels are too close together at a certain point in the future, potentially violating the minimum safe meeting distance, or if multiple vessels simultaneously approach the virtual anchorage area, exacerbating congestion, these conflicts consider not only physical overlap but also temporal overlap, i.e., excessively high vessel density or disordered waiting sequences in the same area at the same time.

[0083] Next, based on the identified potential spatiotemporal conflicts, risk assessment results are generated for different scheduling schemes. Specifically, the system can simulate various scheduling strategies, such as maintaining the current plan, adjusting the port entry sequence, reallocating pilotage resources, or modifying the navigation plan, and quantify key indicators such as congestion risk, average vessel waiting time, overall throughput efficiency, or potential collision probability under each scheme. By comparing the assessment results, it is clear which scheduling adjustments can effectively reduce risks, shorten waiting times, or improve overall traffic flow. This assessment process provides objective data support for decision-making, avoiding the subjectivity of relying on experience-based judgments in traditional scheduling.

[0084] Then, based on the assessment results, specific adjustments were made to the scheduling decision-making scheme for ocean-going vessels. These adjustments primarily included optimizing the scheduling sequence, such as prioritizing the passage of vessels that have been waiting for a long time or preventing vessels with missing signals from entering narrow waterways simultaneously with those that are functioning normally; modifying navigation plans, such as suggesting that some vessels temporarily detour outside the virtual anchorage area or change their speed to avoid peak times; and improving resource allocation strategies, such as dynamically allocating tugboats, pilots, or other port resources to match predicted peak vessel arrival times. These adjustments directly address the hidden problems revealed by the predicted trajectories, ensuring that the scheduling scheme is more closely aligned with real-time maritime traffic conditions.

[0085] As an implementable approach, adjusting the scheduling of ocean-going vessels based on the predicted trajectory to obtain a scheduling decision scheme includes: identifying potential spatiotemporal conflict relationships between multiple ocean-going vessels and the virtual anchorage area and other vessels' predicted trajectories within the predicted time range, based on the predicted trajectory; generating an assessment result reflecting changes in vessel congestion risk or waiting time under different scheduling schemes based on the potential spatiotemporal conflict relationships; and adjusting the vessel scheduling sequence, navigation plan, and resource allocation strategy based on the assessment result to generate a scheduling decision scheme.

[0086] Specifically, firstly, based on the predicted trajectories of all vessels (including the actual or extended trajectories of normal vessels and the predicted trajectories of vessels lacking signals), potential spatiotemporal conflict relationships are identified. In practice, the system sets a prediction time range, such as 1 to 4 hours into the future. Using a simulation step size of 1 minute or 5 minutes, the predicted position sequence of each ocean-going vessel is scanned time-by-time. For each simulation moment, the nearest encounter distance and encounter time between all vessel pairs are calculated, and the number of vessels falling into the virtual anchorage area and its buffer zone at each moment is counted. If the nearest encounter distance between two vessels is less than a safety threshold (e.g., 0.5 nautical miles) and the encounter time is less than a critical value (e.g., 10 minutes), it is marked as a potential collision conflict; if the vessel density or number within the virtual anchorage area exceeds a preset capacity threshold at a certain time (e.g., more than 15 vessels simultaneously anchored), it is marked as a congestion conflict. These conflict relationships are stored in list or matrix form, including conflict type, involved vessels, time period, and severity.

[0087] Secondly, based on the identified potential spatiotemporal conflict relationships, evaluation results are generated under different scheduling schemes. The system pre-generates multiple alternative scheduling schemes, such as maintaining the status quo, adjusting the port entry order (e.g., prioritizing the passage of vessels with the longest waiting time), changing the speed or route of some vessels, and reallocating pilot or tugboat resources. For each scheme, after adjusting the planned arrival time, speed instructions, or waypoints of the corresponding vessels, the aforementioned simulation process is rerun to calculate a new conflict list and quantify key evaluation indicators: congestion risk can be defined as the maximum number of vessels simultaneously staying in the virtual anchorage area or the average density; waiting time can be calculated as the sum or average of the differences between the actual port entry time and the earliest possible port entry time for each vessel; additional auxiliary indicators such as total transit time and fuel consumption estimates can also be added. By comparing the indicator values ​​of multiple schemes, the system generates an evaluation report, ranking the scheme with the lowest risk, shortest waiting time, or highest overall score.

[0088] Finally, based on the evaluation results, specific adjustments are made to the vessel scheduling sequence, navigation plan, and resource allocation strategy, generating a final scheduling decision scheme. The system selects the scheme with the highest evaluation score, or it is confirmed by the dispatcher after minor adjustments based on the report. For example, the original scheduling sequence is changed from first-to-first service to a priority order based on waiting time weighting; speed adjustment instructions are issued to vessels with severe conflicts, causing them to slow down and wait outside the virtual anchorage area to avoid peak hours; limited pilot resources are dynamically allocated to the vessel group expected to arrive safely in port earliest. The adjusted scheme is output in a structured form, including a new scheduling sequence number for each vessel, a suggested navigation plan (waypoints, estimated arrival time, recommended speed range), and a resource allocation table (such as matching pilot numbers with corresponding vessels). This scheme can be directly pushed to the vessel traffic management system or ship-to-shore communication platform for execution.

[0089] Overall, this technology enables intelligent optimization of ocean-going vessel scheduling through predictive trajectory-driven conflict identification, scheme evaluation, and decision adjustment.

[0090] The methods provided in this application can be applied to various scenarios, including but not limited to: First, in busy port waters, when multiple vessels form temporary dense waiting areas due to waiting for berths or pilotage, this method can dynamically generate virtual anchorage areas by mining group collaborative behavior, accurately predict the stopping paths of vessels lacking AIS signals, avoid scheduling chaos caused by insufficient identification of traditional fixed anchorages, optimize port entry order, and reduce average waiting time. Second, in narrow channels or under severe weather conditions, vessels often collectively slow down to avoid each other, forming de facto convergence areas. This method utilizes the stability of group heading and deceleration characteristics to identify implicit constraints, detect potential spatiotemporal conflicts in advance, adjust navigation plans and resource allocation, significantly reduce collision risks, and improve channel traffic efficiency. Finally, in areas with dense fishing vessel operations or around offshore wind farms, the loitering operations of fishing vessels or engineering vessels can easily cause signal anomalies. This method can complete missing trajectories and assess congestion risks, providing intelligent decision support for the ship traffic management system, ensuring the safe and efficient detour or waiting of ocean-going cargo ships, and has significant economic and safety value.

[0091] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0092] According to another embodiment, an intelligent scheduling system for ocean-going vessels based on AIS spatiotemporal trajectory data is provided. Figure 3 A schematic block diagram of an intelligent scheduling system for ocean-going vessels based on AIS spatiotemporal trajectory data, according to one embodiment, is shown. Figure 3 As shown, the system 300 includes:

[0093] The trajectory data acquisition unit 301 is configured to acquire AIS spatiotemporal trajectory data of multiple ships in the target sea area, perform temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data, construct a spatiotemporal correlation structure of multiple ships in the dimensions of heading, speed and arrival time, and construct a ship group collaborative behavior mode based on the spatiotemporal correlation structure.

[0094] The virtual anchorage area generation unit 302 is configured to identify, under the constraints of the ship group cooperative behavior mode, a spatiotemporal area in which ships exhibit convergent deceleration, stop or dense convergence within a predetermined time window, and dynamically generate a virtual anchorage area based on the stability and continuity of the group behavior within the spatiotemporal area.

[0095] The predicted trajectory generation unit 303 is configured to use the virtual anchorage area as an implicit navigation constraint and combine it with the ship group cooperative behavior mode to form a joint constraint, and to predict the potential navigation path of ships with missing or abnormal AIS signals, thereby obtaining the predicted trajectory of the ship.

[0096] The scheduling decision generation unit 304 is configured to adjust the scheduling of ocean-going vessels based on the predicted trajectory to obtain a scheduling decision scheme.

[0097] As an implementable approach, the trajectory data acquisition unit 301, when performing temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data to construct a spatiotemporal correlation structure for multiple ships in the dimensions of heading, speed, and arrival time, can be configured as follows: Based on the transmission frequency and time interval distribution of the AIS spatiotemporal trajectory data of multiple ships within the target sea area, an adaptive time reference is determined, and the AIS spatiotemporal trajectory data of each ship is mapped to a unified temporal representation under the time reference; under the unified temporal representation, a spatiotemporal correlation structure is constructed with ships as nodes and spatial proximity and similarity of navigation states as edge weights, wherein the similarity of navigation states is jointly determined by the trend of heading change, the trend of speed change, and the arrival time offset.

[0098] As an implementable approach, the trajectory data acquisition unit 301, when constructing a ship group collaborative behavior pattern based on the spatiotemporal correlation structure, can be configured to: within a predetermined time window, calculate the correlation between multiple ships in terms of heading change trend, speed change trend, and arrival time offset according to the spatiotemporal correlation structure; identify a set of ships exhibiting synchronous change characteristics within the time window based on the correlation; evaluate the stability of the synchronous change of the ship set within a continuous time window; and determine the ship set that maintains stable synchronous change within the continuous time window as the ship group collaborative behavior pattern.

[0099] As an implementable approach, when the virtual anchorage area generation unit 302 identifies spatiotemporal regions where ships exhibit convergent deceleration, stagnation, or dense convergence within a predetermined time window under the constraints of the ship group's collaborative behavior pattern, it can be configured to: perform joint analysis on the rate of change of speed and the magnitude of change of course of multiple ships participating in the same ship group's collaborative behavior pattern within the predetermined time window, and identify behavioral segments in which multiple ships exhibit synchronous deceleration or synchronous course stabilization within similar time periods; based on the synchronous behavioral segments, determine the spatial overlap region of multiple ships within the corresponding time period; evaluate the persistence of the spatial overlap region within a continuous time window; and determine the spatial overlap region that remains persistent within the continuous time window as the spatiotemporal region exhibiting convergent deceleration, stagnation, or dense convergence.

[0100] As an implementable approach, when the virtual anchorage area generation unit 302 dynamically generates a virtual anchorage area based on the stability and persistence of group behavior within the spatiotemporal region, it can be configured to: statistically analyze the changes in the number of ships participating in the ship group's collaborative behavior pattern over time, as well as the fluctuations in the ships' heading and speed within the spatiotemporal region; calculate a stability index reflecting the stability of group behavior based on the changes and fluctuations in the number of ships over time; analyze the changing trend of the stability index within a continuous time window to determine the persistence of the group behavior; and dynamically determine the spatial range, duration, and effective time period of the virtual anchorage area based on the stability index and persistence results.

[0101] As an implementable approach, the predicted trajectory generation unit 303, when using the virtual anchorage area as an implicit navigation constraint and combining it with the ship group cooperative behavior pattern to form a joint constraint, can be configured to predict the potential navigation path of ships with missing or abnormal AIS signals, and obtain the predicted trajectory of the ship, as follows: Based on the ship group cooperative behavior pattern, determine the typical heading change trend and speed change range of the ships participating in the group cooperative behavior within the corresponding time window; use the virtual anchorage area as a constraint condition for the potential dwelling or convergence area of ​​ships with missing or abnormal AIS signals, limiting the reachable space of the ship within the predicted time range; under the joint constraints of the typical heading change trend, speed change range, and reachable space, infer the potential navigation path of the ship with missing or abnormal AIS signals within the predicted time range; and determine the potential navigation path as the predicted trajectory of the ship.

[0102] As an implementable approach, the scheduling decision generation unit 304, when adjusting the scheduling of ocean-going vessels based on the predicted trajectory to obtain a scheduling decision scheme, can be configured to: identify potential spatiotemporal conflict relationships between multiple ocean-going vessels and the virtual anchorage area and other vessels' predicted trajectories within the predicted time range based on the predicted trajectory; generate an assessment result reflecting changes in vessel congestion risk or waiting time under different scheduling schemes based on the potential spatiotemporal conflict relationships; and adjust the vessel scheduling sequence, navigation plan, and resource allocation strategy based on the assessment result to generate a scheduling decision scheme.

[0103] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. Components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0104] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0105] In addition, embodiments of this application also provide a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0106] And an electronic device comprising: one or more processors; and a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method described in any of the foregoing method embodiments.

[0107] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the method described in any of the foregoing method embodiments.

[0108] in, Figure 4 An exemplary architecture of an electronic device is shown, which may include a processor 410, a video display adapter 411, a disk drive 412, an input / output interface 413, a network interface 414, and a memory 420. The processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420 can communicate with each other via a communication bus 430.

[0109] The processor 410 can be implemented using a general-purpose CPU, microprocessor, application-specific integrated circuit (ASIC), or one or more integrated circuits to execute relevant programs and implement the technical solution provided in this application.

[0110] The memory 420 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 420 can store the operating system 421 for controlling the operation of the electronic device 400, and the basic input / output system (BIOS) 422 for controlling the low-level operations of the electronic device 400. Additionally, it can store a web browser 423, a data storage management system 424, and an intelligent dispatching system for ocean-going vessels based on AIS spatiotemporal trajectory data 425, etc. The aforementioned intelligent dispatching system for ocean-going vessels based on AIS spatiotemporal trajectory data 425 can be the application program that specifically implements the aforementioned steps in this embodiment. In summary, when implementing the technical solution provided in this application through software or firmware, the relevant program code is stored in the memory 420 and executed by the processor 410.

[0111] Input / output interface 413 is used to connect input / output modules to realize information input and output. Input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. Input devices may include keyboards, mice, touch screens, microphones, various sensors, etc., and output devices may include displays, speakers, vibrators, indicator lights, etc.

[0112] Network interface 414 is used to connect a communication module (not shown in the figure) to enable communication and interaction between this device and other devices. The communication module can communicate via wired means (such as USB, Ethernet cable, etc.) or wireless means (such as mobile network, WIFI, Bluetooth, etc.).

[0113] Bus 430 includes a pathway for transmitting information between various components of the device, such as processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, and memory 420.

[0114] It should be noted that although the above-described device only shows the processor 410, video display adapter 411, disk drive 412, input / output interface 413, network interface 414, memory 420, bus 430, etc., in specific implementations, the device may also include other components necessary for normal operation. Furthermore, those skilled in the art will understand that the above-described device may only include the components necessary for implementing the solution of this application, and does not necessarily include all the components shown in the figures.

[0115] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a computer program product. This computer program product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of this application.

[0116] The technical solutions provided in this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for intelligent scheduling of ocean-going vessels based on AIS spatiotemporal trajectory data, characterized in that, The method includes: Acquire AIS spatiotemporal trajectory data of multiple ships within the target sea area, perform temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data, construct a spatiotemporal correlation structure of multiple ships in the dimensions of heading, speed and arrival time, and construct a ship group collaborative behavior mode based on the spatiotemporal correlation structure; Under the constraints of the aforementioned ship group cooperative behavior pattern, identify the spatiotemporal regions in which ships exhibit convergent deceleration, stagnation, or dense convergence within a predetermined time window; Within the spatiotemporal region, the number of ships participating in the collaborative behavior pattern of the ship group is statistically analyzed over time, as well as the fluctuation range of the ships in terms of course and speed. Based on the changes and fluctuations in the number of ships over time, a stability index reflecting the stability of group behavior is calculated. The stability index is analyzed over a continuous time window to determine the persistence of the group behavior. Based on the stability indicators and persistence results, the spatial range, duration, and effective period of the virtual anchorage area are dynamically determined; Based on the ship group cooperative behavior pattern, the typical course change trend and speed change range of the ships participating in the group cooperative behavior within the corresponding time window are determined. The virtual anchorage area is used as a constraint on the potential dwelling or convergence area of ​​ships with missing or abnormal AIS signals, thus limiting the reachable space of the ships within the predicted time range. Under the combined constraints of the typical course change trend, speed change range, and reachable space, the potential navigation path of a ship with missing or abnormal AIS signal is inferred within the predicted time range. The potential navigation path is determined as the predicted trajectory of the vessel; Based on the predicted trajectory, the scheduling of ocean-going vessels is adjusted to obtain a scheduling decision scheme.

2. The method according to claim 1, characterized in that, The step of performing temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data to construct a spatiotemporal correlation structure for multiple ships in the dimensions of heading, speed, and arrival time includes: Based on the transmission frequency and time interval distribution of AIS spatiotemporal trajectory data of multiple ships in the target sea area, an adaptive time reference is determined, and the AIS spatiotemporal trajectory data of each ship is mapped to a unified time series representation under the time reference. Under a unified temporal representation, a spatiotemporal association structure is constructed with ships as nodes and spatial proximity and similarity of navigation status as edge weights, wherein the similarity of navigation status is jointly determined by the trend of heading change, the trend of speed change, and the arrival time offset.

3. The method according to claim 1, characterized in that, The construction of the ship group collaborative behavior pattern based on the spatiotemporal correlation structure includes: Within a predetermined time window, the correlation between multiple ships in terms of course change trend, speed change trend and arrival time offset is calculated according to the spatiotemporal correlation structure. Based on the degree of correlation, identify the set of ships that exhibit synchronous change characteristics within the time window; The stability of the synchronous changes of the ship assembly within a continuous time window is evaluated; A group of ships that maintains stable and synchronous changes within a continuous time window is identified as a ship group cooperative behavior pattern.

4. The method according to claim 1, characterized in that, The identification of spatiotemporal regions where ships exhibit convergent deceleration, stagnation, or dense convergence within a predetermined time window, under the constraints of the ship group's cooperative behavior pattern, includes: Within the predetermined time window, the speed change rate and course change magnitude of multiple ships participating in the same ship group's coordinated behavior pattern are jointly analyzed to identify behavioral segments in which multiple ships simultaneously decelerate or simultaneously stabilize their course within a similar time period. Based on the synchronized behavior segments, the spatial overlap area of ​​multiple ships within the corresponding time period is determined; The persistence of the spatially overlapping region within a continuous time window is evaluated. Spatial overlap regions that remain continuous within a continuous time window are identified as spatiotemporal regions exhibiting convergent deceleration, stagnation, or dense convergence.

5. The method according to claim 1, characterized in that, The adjustment of the scheduling of ocean-going vessels based on the predicted trajectory to obtain a scheduling decision scheme includes: Based on the predicted trajectory, potential spatiotemporal conflict relationships between multiple ocean-going vessels and the virtual anchorage area and other vessel predicted trajectories within the predicted time range are identified. Based on the potential spatiotemporal conflict relationships, an assessment result reflecting the changes in ship congestion risk or waiting time under different scheduling schemes is generated; Based on the evaluation results, the scheduling order, navigation plan and resource allocation strategy of ships are adjusted to generate a scheduling decision scheme.

6. A smart scheduling system for ocean-going vessels based on AIS spatiotemporal trajectory data, characterized in that, The system includes: The trajectory data acquisition unit is configured to acquire AIS spatiotemporal trajectory data of multiple ships in the target sea area, perform temporal alignment and spatial correlation analysis on the AIS spatiotemporal trajectory data, construct a spatiotemporal correlation structure of multiple ships in the dimensions of heading, speed and arrival time, and construct a ship group collaborative behavior mode based on the spatiotemporal correlation structure. The virtual anchorage area generation unit is configured to, under the constraints of the ship group's collaborative behavior pattern, identify spatiotemporal areas where ships exhibit convergent deceleration, lingering, or dense convergence within a predetermined time window; within these spatiotemporal areas, statistically analyze the changes in the number of ships participating in the ship group's collaborative behavior pattern over time, as well as the fluctuations in the ships' heading and speed; based on the changes and fluctuations in the number of ships over time, calculate a stability index reflecting the stability of the group's behavior; analyze the trend of the stability index within a continuous time window to determine the persistence of the group's behavior; and dynamically determine the spatial range, duration, and effective time period of the virtual anchorage area based on the stability index and persistence results. The predicted trajectory generation unit is configured to, based on the ship group cooperative behavior pattern, determine the typical heading change trend and speed change range of the ships participating in the group cooperative behavior within the corresponding time window; use the virtual anchorage area as a constraint condition for the potential dwelling or convergence area of ​​ships with missing or abnormal AIS signals, limiting the reachable space of the ships within the predicted time range; under the joint constraints of the typical heading change trend, speed change range, and reachable space, infer the potential navigation path of the ships with missing or abnormal AIS signals within the predicted time range; and determine the potential navigation path as the predicted trajectory of the ships. The scheduling decision generation unit is configured to adjust the scheduling of ocean-going vessels based on the predicted trajectory to obtain a scheduling decision.

7. An electronic device, characterized in that, include: One or more processors; And a memory associated with the one or more processors, the memory being used to store program instructions that, when read and executed by the one or more processors, perform the steps of the method according to any one of claims 1 to 5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the steps of the method according to any one of claims 1 to 5.

Images