A method and system for identifying abnormal behavior of a ship's approach and departure speed from a port

Abstract

The application provides a kind of ship in and out of port speed abnormal behavior identification method and system, applied to shipping safety monitoring technical field, the method comprises: obtaining ship AIS data point sequence, based on geographic fence and speed threshold, judge whether the geographic location of consecutive data points falls into the fence, if so, judge whether the speed of previous data point is normal and the speed of current data point is abnormal, if meet, generate early warning signal, the application can accurately identify the'mutation process' of speed instead of just 'abnormal state' through the above scheme, solves the technical problem that prior art early warning is not timely, realizes earlier and more accurate early warning to abnormal behavior of ship in and out of port speed.

Description

TECHNICAL FIELD

[0001] A method and system for identifying abnormal behavior of a ship's speed when entering or leaving a port BACKGROUND

[0002] With the rapid development of global shipping industry, the traffic flow of ships in ports and waterways is increasingly dense, and the management of navigation safety and efficiency is put forward with higher requirements. In traffic-intensive or restricted water areas such as ports and waterways, the speed of ships is a parameter that needs to be strictly controlled. Too fast speed may lead to major safety accidents such as collision and grounding; too slow speed will reduce the efficiency of waterway traffic and cause congestion.

[0003] Currently, the maritime monitoring system can monitor the instantaneous speed of a ship using Automatic Identification System (AIS) data. When the speed of a ship exceeds the maximum limit or is lower than the minimum limit, the system can issue an alarm. For example, Chinese patent applications CN116257565A and CN117570997A disclose methods for monitoring the speed of a ship in a specific water area, the logic of which is to determine whether the instantaneous speed value of a ship at a certain time falls outside a pre-set normal interval.

[0004] However, the existing technology has a technical defect: it can only identify that the speed of a ship is in an abnormal state, but cannot identify the process of abnormal change in speed. For example, the speed of a ship may suddenly drop from 10 knots of normal speed to 4 knots below the minimum speed of 5 knots in a short time. The existing technology can only trigger an alarm when the speed drops below 5 knots, which may have missed the opportunity for early warning and intervention. This "sudden process" from normal to abnormal is itself more dangerous and valuable information than the abnormal state result, but the existing technology cannot effectively identify it. This lag in early warning may cause the maritime management department to miss the best opportunity to take intervention measures such as collision avoidance and waterway control, thus failing to effectively reduce the probability of accidents.

[0005] Therefore, there is a need in the art for a technical solution that can accurately identify the process of "sudden speed reduction" or "sudden speed increase" of a ship, thereby achieving earlier risk warning.

[0006] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present application, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0007] In view of this, the present application provides a kind of ship in and out of port speed abnormal behavior identification method and system, to solve the technical problems that only the instantaneous speed of ship can be judged in prior art, and the process behavior of " sudden speed reduction " or " sudden acceleration " of ship speed cannot be effectively identified, by obtaining the continuous AIS time series data of ship, and based on the preset channel geographic fence and navigation rule threshold, using the sliding window containing two consecutive data points, whether the preset state transition of ship speed state occurs is judged, to identify the speed change behavior of ship, to realize the accurate identification and timely warning of ship speed mutation behavior.

[0008] The embodiment of the present application provides a kind of ship in and out of port speed abnormal behavior identification method, comprising:

[0009] A series of time-ordered AIS data points containing target ship geographic position and ground speed are obtained;

[0010] Based on the pre-stored geographic fence for defining ship channel on the map and the speed threshold associated with the geographic fence;

[0011] Select the current data point and the previous data point in time sequence in AIS data point, judge whether the geographic position of current data point and the previous data point falls into the geographic fence;

[0012] In response to the geographic position of current data point and the previous data point falling into the geographic fence, based on the speed threshold, judge whether the ground speed of the previous data point is in the normal speed interval, and whether the ground speed of the current data point is in the abnormal speed interval;

[0013] When the ground speed of the previous data point is in the normal speed interval, and the ground speed of the current data point is in the abnormal speed interval, a warning signal is generated to indicate that the ship has speed abnormal behavior.

[0014] In some optional embodiments, the abnormal speed interval is lower than the preset minimum speed threshold.

[0015] In some optional embodiments, the abnormal speed interval is higher than the preset maximum speed threshold.

[0016] In some optional embodiments, the method further comprises:

[0017] After generating the warning signal, continue to obtain and judge the ground speed of subsequent AIS data points, and when it is detected that the ground speed returns to the normal speed interval from the abnormal speed interval, an end signal is generated to indicate the end of speed abnormal behavior.

[0018] In some optional embodiments, the method further comprises:

[0019] After the step of judging whether the geographical positions of the current data point and the previous data point both fall within the geo-fence, when it is judged that the geographical position of the previous data point is outside the geo-fence and the geographical position of the current data point is within the geo-fence, the step of generating the early warning signal is not performed.

[0020] In some optional embodiments, the method further comprises:

[0021] Before the step of generating the early warning signal, the method further comprises:

[0022] Based on the pre-stored acceleration threshold associated with the geo-fence, the rate of change of the airspeed between the current data point and the previous data point is calculated, and it is judged whether the absolute value of the rate of change of the airspeed is greater than the acceleration threshold;

[0023] The early warning signal is generated when the absolute value of the rate of change of the airspeed is greater than the acceleration threshold.

[0024] In some optional embodiments, the step of calculating the rate of change of the airspeed comprises:

[0025] The respective time stamps of the current data point and the previous data point are extracted respectively;

[0026] The time difference is calculated based on the time stamps;

[0027] The difference in airspeed of the current data point and the previous data point is divided by the time difference to obtain the rate of change of the airspeed.

[0028] In some optional embodiments, the pre-stored geo-fence is divided into a plurality of sub-regions of the airway, and a set of airspeed thresholds is independently associated with each sub-region of the airway.

[0029] In some optional embodiments, the method further comprises:

[0030] Before the step of judging whether the geographical positions of the current data point and the previous data point both fall within the geo-fence, first determine the sub-region of the airway to which the current data point and the previous data point both belong according to the geographical positions of the current data point and the previous data point, and call the set of airspeed thresholds associated with the sub-region of the airway for subsequent judgment.

[0031] In some optional embodiments, the step of generating the end signal specifically comprises:

[0032] When it is detected that the airspeed over the ground returns to the normal airspeed interval from the abnormal airspeed interval, it is further judged whether the airspeed over the ground of the subsequent at least one AIS data point continuously remains within the normal airspeed interval;

[0033] After it is determined that the airspeed over the ground continuously remains within the normal airspeed interval, the end signal is generated.

[0034] In some alternative embodiments, the method further comprises:

[0035] When it is determined that the geographic position of the previous data point is within the geo-fence and the geographic position of the current data point is outside the geo-fence, if there is a generated early warning signal but no end signal generated for the abnormal speed behavior, an end signal is generated.

[0036] In some alternative embodiments, the pre-stored geo-fence is divided into a plurality of sub-regions of the shipping lane, and a set of speed threshold values and a set of acceleration threshold values are independently associated with each sub-region of the shipping lane.

[0037] In some alternative embodiments, the method further comprises:

[0038] Before the step of determining whether the geographic positions of the current data point and the previous data point are both within the geo-fence, the sub-region of the shipping lane to which the current data point and the previous data point both belong is determined according to the geographic positions of the current data point and the previous data point, and the set of speed threshold values and the set of acceleration threshold values associated with the sub-region of the shipping lane are called for subsequent determination.

[0039] In some alternative embodiments, the obtained series of AIS data points are continuous data streams from the same target ship.

[0040] In some alternative embodiments, the generated early warning signal contains the timestamp, the geographic position, the speed over ground, and the type of abnormal speed interval of the current data point.

[0041] Embodiments of the present application provide a system for identifying abnormal speed behavior of a ship entering or leaving a port, comprising:

[0042] a data acquisition module configured to obtain a series of time-ordered automatic identification system (AIS) data points containing the geographic position and the speed over ground of a target ship;

[0043] a storage module configured to store a geo-fence for defining a shipping lane on a map and speed threshold values associated with the geo-fence;

[0044] a processor communicatively connected to the data acquisition module and the storage module, the processor being configured to:

[0045] select a current data point and a previous data point in the series of AIS data points that are continuous in time, and determine whether the geographic positions of the current data point and the previous data point are both within the geo-fence stored in the storage module;

[0046] in response to the geographic positions of the current data point and the previous data point both being within the geo-fence, determine whether the speed over ground of the previous data point is within a normal speed interval and the speed over ground of the current data point is within an abnormal speed interval based on the speed threshold values;

[0047] a signal generation module, communicatively connected to the processor, and configured to:

[0048] generate a warning signal indicating that the abnormal speed behavior of the ship has occurred when the processor determines that the ground speed of the previous data point is in the normal speed interval and the ground speed of the current data point is in the abnormal speed interval.

[0049] In some optional embodiments, the processor is further configured to:

[0050] After the signal generation module generates the warning signal, the ground speed of the subsequent AIS data point is continuously acquired and judged;

[0051] The signal generation module is further configured to:

[0052] generate an end signal indicating that the abnormal speed behavior has ended when the processor detects that the ground speed has returned to the normal speed interval from the abnormal speed interval.

[0053] In some optional embodiments, the storage module is further configured to store an acceleration threshold value associated with the geofence;

[0054] The processor is further configured to:

[0055] After determining the interval in which the ground speed is located, the speed change rate between the current data point and the previous data point is calculated, and it is determined whether the absolute value of the speed change rate is greater than the acceleration threshold value;

[0056] The signal generation module is specifically configured to:

[0057] generate a warning signal when the processor determines that the absolute value of the speed change rate is greater than the acceleration threshold value.

[0058] In some optional embodiments, the geofence stored in the storage module is divided into a plurality of channel sub-regions, and a set of speed threshold values is independently associated with each channel sub-region;

[0059] The processor is further configured to:

[0060] Before determining whether the geographic positions of the current data point and the previous data point are both within the geofence, first determine the channel sub-region to which they both belong according to the geographic positions of the current data point and the previous data point, and call the set of speed threshold values associated with the channel sub-region for subsequent determination.

[0061] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application.

[0062] The ship speed abnormal behavior identification method and system have the following advantages:

[0063] The present application can early warn abnormal behavior of ship speed in and out of port by identifying the mutation process of ship speed from normal to abnormal instead of only monitoring abnormal state, so as to strive for more sufficient intervention time. At the same time, since the mutation process of ship speed is considered, by quantifying the ship speed change rate and judging whether it exceeds the preset acceleration threshold, the dangerous "mutation" and the normal "slow change" can be effectively distinguished, the false alarm caused by the ship normal speed change crossing the threshold in the prior art is significantly reduced, the warning information is more valuable, the accuracy of the warning is improved, and the navigation safety guarantee ability of the port channel is enhanced. BRIEF DESCRIPTION OF DRAWINGS

[0064] Other features, objects, and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0065] Figure 1 is a general flowchart of a ship speed abnormal behavior identification method provided by an embodiment of the present application;

[0066] Figure 2 is a flowchart of a ship speed abnormal behavior identification method of an embodiment of the present application;

[0067] Figure 3 is a schematic diagram of ship navigation trajectory and sliding window working principle in an embodiment of the present application;

[0068] Figure 4 is a time axis schematic diagram of a ship speed abnormal behavior identification process in a specific embodiment of the present application;

[0069] Figure 5 is a structural schematic diagram of a ship speed abnormal behavior identification system of an embodiment of the present application. DETAILED DESCRIPTION

[0070] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the implementations set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example implementations to those skilled in the art. The described features, structures, or characteristics can be combined in one or more implementations.

[0071] Furthermore, the accompanying drawings are included to provide a further understanding of the present application, and are incorporated in and constitute a part of this specification. The drawings illustrate embodiments of the present application and, together with the description, serve to explain the principles of the present application. In the drawings:

[0072] The flowcharts shown in the drawings are only illustrative, and do not necessarily include all the steps. For example, some steps can be further divided, and some steps can be combined or partially combined, so the actual execution order can be changed according to the actual situation.

[0073] The automatic identification system (AIS) periodically broadcasts the dynamic information of the ship through radio communication technology, including the unique identification, geographic position, speed, etc. of the ship. According to the AIS data, the navigation state of the ship can be monitored. The navigation behavior of the ship in the water area is affected by many factors, such as channel restrictions, traffic density, weather conditions, etc. The speed as a parameter of the ship navigation state, its change directly reflects the ship's maneuvering behavior and potential risk. Through the analysis of the historical AIS data of the ship, a statistical model of the normal navigation state of the ship can be established. When the actual speed of the ship deviates from the normal range, it may indicate the occurrence of potential danger or abnormal situation. Using the geographic fence technology, the activity range of the ship can be limited, when the ship enters or leaves a specific area, the corresponding event is triggered. Combined with the speed threshold, it can be judged whether the speed of the ship in the specific area meets the safety requirements.

[0074] To make the objectives, technical solutions and advantages of the present application clearer, the specific embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0075] Figure 1 is the overall flowchart of the ship speed abnormal behavior identification method provided by an embodiment of the present application. Referring to Figure 1 , the method described in the present application can include the following steps:

[0076] First, the step of defining the approach channel and the speed threshold in the channel is performed. This step provides the basis rules and geographical boundaries for the subsequent identification and judgment. In a specific implementation, this step includes defining one or more closed polygons as a geographical fence on an electronic map through a series of latitude and longitude coordinate points to limit the monitoring channel of the ship; at the same time, a set or multiple sets of speed rules are associated with each geographical fence, which at least includes a speed threshold for defining the normal speed range (for example, the minimum speed threshold and the maximum speed threshold). These geographical fence coordinates and rule parameters are pre-stored in the storage module of the system.

[0077] Next, the step of collecting real-time AIS data of the ship is performed. The data collection module of the system continuously and real-time acquires the AIS message broadcast by the target ship through the AIS receiving device or network interface, and parses the data points containing the time stamp, geographical position and ground speed from the AIS message to form a time-ordered data point sequence.

[0078] Then, the application introduces a time series analysis logic based on continuous data points to identify the "process" of speed change rather than just the "state". Specifically, the system uses a sliding window of size 2, which always contains the current data point and the previous data point in time. The system compares the current point speed with the speed threshold to determine the speed state at the current time; at the same time, the system also compares the previous point speed with the speed threshold to determine the speed state at the previous time. The two comparison steps are the core of the application, which aims to obtain the speed state information of the two consecutive time points to provide the basis for the subsequent behavior judgment.

[0079] After the above comparison is completed, the step of judging whether there is a sudden speed reduction / acceleration is performed. The core task of this step is to analyze the speed state change between the two consecutive data points. The basic judgment logic of this step is to judge whether the speed state of the ship has a preset transition from the normal speed range to the abnormal speed range between the previous data point and the current data point.

[0080] Finally, if the judgment result is yes, i.e., it is determined that the preset state transition occurs, the step of notifying the relevant personnel when the speed is abnormal is performed. This step is usually performed by the signal generation module of the system, for example, generating a structured warning signal containing information such as event time, position, speed, etc., and sending it to the maritime monitoring personnel through the user interface, sound and light alarm or network interface.

[0081] It should be emphasized that, Figure 1 The figure shows the overall flow of the application. In specific application scenarios, in order to obtain better technical effects, the above steps can include more refined technical solutions.

[0082] For example, in a preferred embodiment, the step of "judging whether there is sudden speed reduction / acceleration" can further include a quantitative judgment on the rate of change of speed, so as to enhance the accuracy of recognition. Specifically, after judging that the speed state has shifted, the processor will further calculate the actual speed change rate (i.e. acceleration) between the current data point and the previous data point according to their timestamps and ground speed. Then, the processor will compare the absolute value of the speed change rate with a preset acceleration threshold. Only when both the "state shift" and "change rate exceeding the acceleration threshold" conditions are met, the system will finally determine that there is an abnormal behavior of sudden speed reduction or acceleration and generate a warning signal. This scheme can effectively distinguish between dangerous speed mutations and slow speed changes caused by normal ship operation by quantifying the severity of the mutation, thereby significantly reducing the false positive rate and improving the value of the warning information.

[0083] For another example, in another preferred embodiment, the step of "defining the approach channel and the speed threshold in the channel" can achieve dynamic and fine management. Specifically, a complete channel can be divided into multiple channel sub-regions with different navigation characteristics (such as open area, narrow area, turning area, etc.), and a set of speed threshold and acceleration threshold suitable for each channel sub-region is configured independently. In operation, the system will first determine the channel sub-region to which the ship currently belongs according to its geographical position, and dynamically call the specific rule parameters associated with the sub-region for judgment. This scheme enables the recognition model to adapt to the complex and variable physical environment of the channel, further improving the accuracy and practicality of recognition.

[0084] As shown in Figure 2 The embodiment of the present application provides a ship speed abnormal behavior recognition method for ships entering and leaving a port, and the method comprises the following steps:

[0085] Step S100: acquiring a series of time-ordered automatic identification system (AIS) data points containing the geographical position and ground speed of a target ship.

[0086] In this step, AIS data is acquired by an AIS receiving device, which can be a shore-based AIS base station or a shipborne AIS transponder. An AIS data point contains at least the unique identification of the ship (MMSI), a time stamp, a geographic position (latitude and longitude), and a speed over ground (SOG). AIS data is transmitted in the form of messages and arranged in chronological order to form a sequence of data points, for example {P1, P2,..., Pk,...}, where each Pk represents an AIS data point. The AIS data can be acquired in various ways, including but not limited to receiving through radio frequency, receiving through network data interface, etc. In other optional embodiments, the data can also be acquired by a satellite AIS receiving device to extend the monitoring range.

[0087] Step S200: Define the monitoring area and judgment criteria based on the pre-stored geographic fence used to define the ship's route on the map and the speed threshold associated with the geographic fence.

[0088] The geographic fence refers to a closed area pre-defined on the electronic map, used to define the ship's navigation area. The geographic fence is composed of a series of ordered latitude and longitude coordinate points, which are stored in the data storage to define the range of a specific route. The speed threshold refers to one or more values associated with the geographic fence, used to determine whether the ship's speed is within the normal range. The speed threshold can include a minimum speed threshold and a maximum speed threshold. These thresholds are set according to the specific circumstances of the route and stored in the data storage in association with the corresponding geographic fence. In other optional embodiments, the geographic fence can also be defined in other forms, such as using a polygon object in the geographic information system (GIS). The speed threshold can also be a dynamic threshold adjusted according to time, weather, etc.

[0089] Step S300: Select the current data point and the previous data point in the AIS data point sequence in time, and determine whether the geographic positions of the current data point and the previous data point fall within the geographic fence.

[0090] In this step, the system uses a sliding window of size 2, which contains two adjacent AIS data points in time: the current data point (Pk) and the previous data point (Pk-1). By calculating the relationship between the latitude and longitude coordinates of the ship and the boundary coordinates of the geographic fence, it is determined whether the data point is located within the geographic fence. The judgment algorithm includes but is not limited to the ray method, the point-in-polygon algorithm, etc. Only when both data points in the sliding window are located within the geographic fence, the subsequent speed judgment is performed. In other optional embodiments, the size of the sliding window can be adjusted according to actual needs, for example, a sliding window containing three or more data points can be used.

[0091] Step S400: in response to the fact that the geographic locations of the current data point and the previous data point both fall within the geofence, determining whether the ground speed of the previous data point is in the normal speed interval and the ground speed of the current data point is in the abnormal speed interval based on the speed threshold.

[0092] In this step, first, the speed threshold associated with the current geofence is read from the memory. Then, it is determined whether the ground speed of the previous data point and the current data point is in the normal speed interval, respectively. The normal speed interval is determined by the minimum speed threshold and the maximum speed threshold. If the ground speed of the previous data point is higher than the minimum speed threshold and lower than the maximum speed threshold, it is considered to be in the normal speed interval. If the ground speed of the current data point is lower than the minimum speed threshold or higher than the maximum speed threshold, it is considered to be in the abnormal speed interval. In some other optional embodiments, the definition of the normal speed interval can be more complex, for example, considering the channel width, water depth, etc.

[0093] Step S500: when the ground speed of the previous data point is in the normal speed interval and the ground speed of the current data point is in the abnormal speed interval, a warning signal indicating that the ship has abnormal speed behavior is generated.

[0094] In this step, if the judgment result of step S400 shows that the ground speed of the previous data point is in the normal speed interval and the ground speed of the current data point is in the abnormal speed interval, the system generates a warning signal. The warning signal is used to indicate that the ship has abnormal speed behavior, such as sudden speed reduction or sudden acceleration. The warning signal can contain the MMSI of the ship, the current timestamp, the geographic location, the ground speed, and the type of abnormality, etc. The warning signal can be displayed through the display, the alarm sound can be emitted through the speaker, or the network interface can be sent to other systems. In some other optional embodiments, the warning signal can also contain other information, such as the recommended measures.

[0095] Through the above steps, the identification of the ship's speed abnormal behavior when entering and leaving the port is realized. The data acquisition module acquires AIS data, the storage module stores the geofence and the speed threshold, the processor executes the data processing and judgment logic, and the signal generation module generates the warning signal. These modules work together to solve the technical problem that the prior art cannot effectively identify the process behavior of the ship's speed "sudden speed reduction" or "sudden acceleration", realize the transition from "state monitoring" to "behavior identification", and can more accurately and earlier warn the ship's speed abnormal behavior when entering and leaving the port.

[0096] Through the above scheme, the ship's speed abnormal behavior when entering and leaving the port can be effectively identified, the early warning of the speed abnormal behavior is realized, and the navigation safety guarantee capability of the port channel is enhanced.

[0097] To illustrate the working principle of the present application, especially the sliding window mechanism and the handling of boundary conditions, the following will be described in detail with reference to Figure 3 the accompanying drawings.

[0098] As shown in Figure 3 , two vertical lines in the figure define the geofencing boundary of a shipping lane. The arrow indicates the direction of the ship's voyage. P1 to P6 represent six AIS data points successively acquired by the system in time sequence, each of which represents the geographical position of the ship at a certain time.

[0099] In this embodiment, the system uses a sliding window of size 2 to analyze the data point sequence. The following will describe the execution details of the method of the present application in the process of moving the sliding window in time sequence:

[0100] Time T1: The system only acquires data point P1. At this time, since the data points are insufficient to constitute a complete sliding window, the system does not make any judgment.

[0101] Time T2: The system acquires data point P2, at which time the first sliding window is formed, containing (P1, P2). The processor judges that the geographical position of P1 is outside the shipping lane, while the geographical position of P2 is inside the shipping lane. This belongs to the boundary condition of the ship entering the shipping lane. According to the rules of the present application, in order to avoid false positives due to normal deceleration when entering the port, the system does not perform the judgment of abnormal behavior of the speed in this sliding window.

[0102] Time T3: The system acquires data point P3, and the sliding window moves forward, containing (P2, P3). As shown in the dashed box in the figure. The processor judges that the geographical positions of P2 and P3 are both inside the shipping lane. At this time, the system enters the regular judgment process. The system will acquire the ground speed of P2 and P3 respectively, and judge their speed states according to the preset speed threshold. Assuming that the speed of P2 is normal, while the speed of P3 is abnormal (for example, lower than the minimum speed), the condition of state transition is met. In a preferred embodiment, the system will further calculate the speed change rate between (P2, P3) and compare it with the acceleration threshold. If all conditions are met, the system will generate a warning signal at time T3.

[0103] Times T4 and T5: Similarly, the system will process the sliding windows (P3, P4) and (P4, P5) in turn. Since all data points in these two windows are inside the shipping lane, the system will continue to perform the regular abnormal behavior judgment logic of the speed.

[0104] Time T6: The system acquires data point P6, and the sliding window moves to contain (P5, P6). The processor determines that the geographic position of P5 is within the channel, while the geographic position of P6 is outside the channel. This belongs to the case of the ship leaving the boundary of the channel. According to the rules of the present application, the system will check whether there is a speed anomaly behavior associated with the ship that has not ended. If there is (for example, the warning event generated at time T3 has not ended), the system will generate an end signal at this moment to ensure the completeness of the event record.

[0105] Through the above detailed description with reference to Figure 3 , the working principle of the present application for analyzing continuous data points by using a sliding window and finely processing different scenarios such as entering, sailing in, and leaving the channel is clearly presented.

[0106] In combination Figure 4 with a scenario containing specific example data, the whole process of identifying abnormal ship speed behavior is described in detail.

[0107] Referring to Figure 4 , the figure is a time axis diagram of a ship speed anomaly behavior identification process. In the figure, the dashed line frame starting from time T4 and ending at time T14 represents a preset geographical range of a monitored channel. The sailing rules set for the channel are pre-stored in the storage module of the system, and are as follows:

[0108] Minimum speed threshold : set to 5.0 knots. Therefore, Figure 4 , the AIS data points represented by solid circles (such as P4, P5) have a ground speed greater than or equal to 5.0 knots and are in the "normal speed interval"; the AIS data points represented by hollow circles (such as P6, P9) have a ground speed less than 5.0 knots and are in the "abnormal speed interval".

[0109] Acceleration threshold : set to 0.8 knots / minute.

[0110] For ease of illustration, it is assumed that the time interval between each adjacent time point from T1 to T15 is 1 minute.

[0111] System execution flow analysis:

[0112] This embodiment will describe how the system processes the AIS data point sequence from P1 to P15 by using a sliding window in a time sequence.

[0113] Time T1-T3: The system receives data points P1, P2, P3 in sequence. When processing the sliding windows (P1, P2) and (P2, P3), the processor determines that the geographical locations of these data points are all outside the preset channel, and thus does not perform the judgment of abnormal behavior of the speed.

[0114] Time T4: The system receives data point P4, forming a sliding window (P3, P4). The processor determines that P3 is outside the channel, and P4 is inside the channel. This case is identified as a boundary condition of “entering the channel”, and in order to avoid false positives due to normal deceleration operation of the ship entering the port, the system does not perform abnormal behavior judgment in this sliding window.

[0115] Time T5: The system receives data point P5, forming a sliding window (P4, P5). The processor determines that P4 and P5 are both inside the channel, and according to the figure, their speed states are both “normal” (solid circle). Since there is no state transition from normal to abnormal, the system does not generate a warning signal.

[0116] Time T6: The system receives data point P6, forming a sliding window (P5, P6). The processor determines that P5 and P6 are both inside the channel. The speed state of P5 is “normal”, and the speed state of P6 is “abnormal” (hollow circle), which meets the condition of state transition. In order to further confirm whether it is “sudden change”, the system performs the acceleration judgment in the preferred solution: assuming that the ground speed of P5 is 5.6 knots and the ground speed of P6 is 4.3 knots, then the absolute value of the speed change rate is |4.3 - 5.6| / 1 = 1.3 knots / minute. This value is greater than the preset acceleration threshold of 0.8 knots / minute. Since both the state transition and the change rate exceed the limit, the signal generation module generates a warning signal indicating “sudden deceleration” at time T6.

[0117] Time T7: The system receives data point P7, forming a sliding window (P6, P7). The speed state of P6 is “abnormal”, and the speed state of P7 returns to “normal”. The system detects that the speed returns from the abnormal speed interval to the normal speed interval, and thus generates an end signal marking that the abnormal behavior starting from time T6 has ended.

[0118] Time T8: The system processes the sliding window (P7, P8), and the speed states of the two data points are both “normal”, and no warning is generated.

[0119] Time T9: The system receives data point P9, forming a sliding window (P8, P9). P8 is "normal" and P9 is "abnormal", satisfying the state transition condition. Assuming P8's ground speed is 5.5 knots and P9's ground speed is 3.2 knots, the absolute value of the speed change rate is |3.2-5.5| / 1 = 2.3 knots / min, which is greater than the acceleration threshold 0.8. The double condition is satisfied, and the system generates a new "sudden speed drop" warning signal at time T9.

[0120] Time T10-T11: The system processes sliding windows (P9, P10) and (P10, P11) in turn. In these windows, the speed state of the previous data point is "abnormal", which does not satisfy the "from normal to abnormal" state transition condition, so the system does not generate a new warning signal, only maintaining the abnormal event started at T9 as active.

[0121] Time T12: The system receives data point P12, forming a sliding window (P11, P12). P11 is "abnormal" and P12 returns to "normal". The system generates an end signal to mark the end of the abnormal behavior starting at T9.

[0122] Time T13: The system processes sliding window (P12, P13), both points are "normal", no warning is generated.

[0123] Time T14: The system receives data point P14, forming a sliding window (P13, P14). P13 is "normal" and P14 is "abnormal", satisfying the state transition. Assuming P13's speed is 5.4 knots and P14's speed is 4.5 knots, the acceleration absolute value is 0.9, which is greater than 0.8. The double condition is satisfied, and the system generates a warning signal at time T14.

[0124] Time T15: The system receives data point P15, forming a sliding window (P14, P15). The processor determines that P14 is within the channel, while P15 is outside the channel, which is a boundary case of "leaving the channel". The system detects that the abnormal behavior started at T14 has not ended, so it forcibly generates an end signal at this moment to ensure the completeness of the event record.

[0125] Through the above detailed step-by-step demonstration combined with Figure 4 , the present application can accurately identify three dangerous speed mutation behaviors at times T6, T9 and T14, and accurately confirm the end of the abnormal state at times T7, T12 and T15. This fully demonstrates that the present application has significant technical progress in the timeliness, accuracy and logical integrity of the warning compared with the prior art.

[0126] In a specific embodiment, on the basis of the above-mentioned embodiment, when the ship's speed state judgment module determines that the ship's speed over ground is lower than the preset minimum speed threshold, the speed state is defined as "too low" or "abnormal". First, the system reads the preset minimum speed threshold from the storage module , which is pre-configured by the administrator and is associated with a specific geo-fence or channel sub-region. Specifically, the speed state judgment module compares the speed over ground SOG in the current AIS data point provided by the data acquisition module with . Then, if SOG is less than , it is determined that the speed state of the data point represented is "too low", and the data point is marked as an abnormal state. Wherein: SOG represents the speed over ground of the current AIS data point; represents the preset minimum speed threshold. Then, the state information is passed to the subsequent abnormal behavior recognition module for determining whether to trigger a warning signal. In some other optional embodiments, the minimum speed threshold may be dynamically adjusted according to the time period, for example, appropriately reducing the value of at night or in bad weather conditions to adapt to different navigation conditions.

[0127] Through the above-mentioned scheme, the embodiment can accurately identify the abnormal state of the ship below the minimum speed threshold, and provide accurate state basis for subsequent warning judgment.

[0128] In a specific embodiment, on the basis of the above-mentioned embodiment, the abnormal speed interval is defined as the speed over ground of the ship being higher than the preset maximum speed threshold. First, the system pre-stores a maximum speed threshold in the storage module for a specific geo-fence or channel sub-region. Specifically, when judging the speed state of the ship, the system compares the speed over ground of the current AIS data point with the maximum speed threshold . The comparison process follows the following judgment formula:

[0129]

[0130] Wherein: represents the speed over ground of the current AIS data point, represents the preset maximum speed threshold. Then, if the above-mentioned judgment formula is established, it is determined that the speed state of the data point is "too high", i.e. in the abnormal speed interval. Next, if the speed state of the previous data point is "normal" (i.e. its speed over ground is not greater than ), and the speed state of the current data point ​the system identifies a state transition from normal to abnormal, and can further combine other conditions (such as the rate of change of the speed) to determine whether to generate a warning signal indicating the behavior of "sudden acceleration". In some optional embodiments, the maximum speed threshold may not be a fixed value, but a dynamic threshold associated with the type of ship, static or dynamic information such as the depth of the ship may be set to 12 knots, while the of a small passenger ship can be set to 18 knots.

[0131] Through the above scheme, the embodiment can accurately identify the abnormal state of the ship speed exceeding the maximum speed limit, and provides a technical basis for effectively avoiding safety risks caused by over-speed driving of the ship.

[0132] In a specific embodiment, on the basis of the above embodiment, after generating a warning signal for indicating that the ship has an abnormal speed behavior, the system continues to monitor the speed state of the ship. Specifically, the system continues to obtain subsequent AIS data points, and determines whether the ground speed of these data points is in the abnormal speed interval based on the preset geofence and speed threshold. Then, the system detects whether the ground speed has a transition from the abnormal speed interval to the normal speed interval. When detecting such a transition, the system generates an end signal for indicating the end of the speed abnormal behavior. In other optional embodiments, the system can also trigger the generation of the end signal by determining whether the ship has exited the preset geofence area, or by receiving a manual intervention instruction.

[0133] Through the above scheme, the embodiment can timely and accurately determine the end time of the speed abnormal behavior of the ship entering and leaving the port, avoid the continuation of the warning state, and enable the monitoring system to more comprehensively reflect the actual sailing state of the ship.

[0134] In one specific embodiment, based on the above embodiment, first, after the system determines whether the geographic locations of the current data point and the previous data point are both within the geographic fence, the system determines the positional relationship between the previous data point and the current data point. Specifically, it is determined whether the geographic location of the previous data point is outside the geographic fence and the geographic location of the current data point is inside the geographic fence. If the result is yes, it indicates that the ship is entering the channel. Then, the system will not perform the step of generating a warning signal, even if the result of the judgment according to the speed threshold satisfies the warning condition. In this way, false alarms caused by normal deceleration of the ship entering the port can be avoided. In other optional embodiments, different geographic fence algorithms (for example, judging the distance between the data point and the geographic fence instead of the absolute inside-outside relationship) can be used to define the state of the ship "entering" the channel, and a configurable port entry mode switch can be set to apply this rule only when the port entry mode is on.

[0135] Through the above scheme, the embodiment can avoid false alarms caused by normal deceleration when the ship enters the monitored channel, and improve the accuracy and reliability of the warning system.

[0136] In one specific embodiment, based on the above embodiment, first, the system obtains a pre-set acceleration threshold associated with the target geographic fence from the storage module. The acceleration threshold represents the speed limit that is considered to be "sudden" acceleration or deceleration in a specific channel. Specifically, the threshold can be obtained by statistical analysis of historical AIS data, or can be set by manual according to the environmental characteristics and safety requirements of the specific channel. Then, before determining the speed state of the current data point and the previous data point, the system calculates the speed change rate between the two data points. The change rate is obtained by analyzing the speed difference between the two data points in time, which reflects the degree of change of the ship's speed per unit time. Then, the system determines whether the absolute value of the calculated speed change rate is greater than the pre-set acceleration threshold. If the absolute value of the speed change rate is less than or equal to the threshold, it is considered that the speed change of the ship is gentle and normal, and it is not an abnormal behavior that needs to be warned. Only when the absolute value of the speed change rate is greater than the threshold, it is considered that the ship has a "sudden" acceleration or deceleration behavior, and combined with the judgment result of the speed state, it is finally determined whether to generate a warning signal. In other optional embodiments, the acceleration threshold can be dynamically adjusted according to different ship types, water depth or weather conditions to adapt to more complex navigation environments.

[0137] By the above scheme, the embodiment can more accurately identify the abnormal sudden change behavior of the ship speed. Compared with the scheme of only relying on the speed threshold, the embodiment not only considers the state of the speed itself, but also considers the degree of change of the speed, so as to effectively filter out the slow speed change caused by normal operation or environmental factors, significantly reduce the false positive rate, and improve the practicability and reliability of the early warning system.

[0138] In a specific implementation, on the basis of the above embodiment, the calculation process of the speed change rate is performed by the processor 30, specifically including: first, extracting the respective time stamps and from the current data point and the previous data point respectively. In a specific implementation, the time stamp is a unified UTC time or Unix time stamp, so as to ensure the accuracy of time calculation. Then, the processor 30 divides the difference between the ground speeds of the current data point and the previous data point by the time difference between them, so as to obtain the accurate speed change rate. The speed change rate A is calculated by the following formula:

[0139]

[0140] Wherein, A represents the speed change rate, which can be in knots / second or knots / minute; represents the ground speed of the current data point ; represents the ground speed of the previous data point ; represents the time stamp of the current data point ; represents the time stamp of the previous data point .

[0141] By dividing the difference between the ground speeds by the actual time difference , the embodiment can adapt to the actual situation of uneven AIS data reporting interval, and ensure the accuracy of the speed change rate calculation. The absolute value of the calculated speed change rate A can be used for comparison with the preset acceleration threshold.

[0142] By the above scheme, the embodiment can accurately calculate the instantaneous speed change rate of the ship, and provide a reliable technical basis for accurately identifying the sudden change process of the ship speed.

[0143] In a specific embodiment, on the basis of the above-mentioned embodiments, the waterway is pre-divided into a plurality of sub-regions which are geographically non-overlapping. Specifically, the division of the waterway can include but is not limited to factors such as the width of the waterway, the depth of the water, the curvature, and the historical accident rate, etc. For example, the area where ships frequently converge in the waterway, or the area with shallow water depth, can be divided into independent sub-regions. Each sub-region corresponds to a unique identifier, such as sub-region A, sub-region B, etc. Then, a set of speed threshold values is independently associated with each sub-region. The speed threshold values include the maximum speed threshold value and / or the minimum speed threshold value. For example, in the sub-region where ships frequently converge, a lower maximum speed threshold value can be set to reduce the risk of collision; while in the sub-region with shallow water depth, a higher minimum speed threshold value can be set to avoid ship grounding. The correspondence between the sub-regions and the speed threshold values can be stored in a database or a configuration file. In some optional embodiments, the waterway can also be divided according to time periods, such as different speed threshold values for daytime and nighttime. Alternatively, the sub-regions do not necessarily overlap geographically, but can partially overlap, and each sub-region corresponds to different navigation rules.

[0144] Through the above scheme, the embodiment can set different speed threshold values according to the characteristics of different regions of the waterway, thereby improving the accuracy of the identification of abnormal behavior of ships entering and leaving the port and reducing the false positive rate.

[0145] In a specific embodiment, on the basis of the above-mentioned embodiments, the method further comprises: first, before the step of judging whether the geographic locations of the current data point and the previous data point both fall within the geographic fence, the system determines the common sub-region of the channel according to the geographic locations of the current data point and the previous data point through a geographic space query. Specifically, the storage module has pre-stored the geographic fence coordinate set of the sub-regions of the channel, and the processor respectively takes the latitude and longitude coordinates of the current data point and the previous data point as input to perform a spatial inclusion query, i.e., to judge whether the coordinates fall within the geographic fence of a sub-region. If both data points only fall within the geographic fence of a unique sub-region, the sub-region is determined as their common attribution. If the two data points fall within the overlapping region of multiple sub-regions, the final attribution sub-region is determined according to the preset priority rules (for example, the priority of the sub-region close to the port is high). Then, the system calls the set of speed threshold values associated with the sub-region of the channel from the storage module for subsequent speed state judgment. For example, if it is determined that the current data point and the previous data point are both located in the "speed limit zone" of the channel entrance, the lower maximum speed threshold value corresponding to the region is called; if it is determined that they are both located in the "smooth zone" of the main channel, the higher maximum speed threshold value corresponding to the region is called, thereby realizing dynamic speed monitoring in different regions. In other optional embodiments, the geographic fence of the sub-region of the channel can be represented in different ways, for example, using a center point and a radius to represent a circular region, or using a grid map to represent an irregular region; the spatial inclusion query can be implemented by different algorithms, for example, the ray method, the radian method or the grid index method.

[0146] Through the above scheme, the present embodiment can dynamically adjust the speed threshold value according to the sub-region of the channel where the ship is located, more finely evaluate the speed state of the ship, improve the accuracy of the early warning, and reduce the false alarm rate.

[0147] In one specific implementation, based on the above embodiment, when the ground speed is detected to recover from the abnormal speed interval to the normal speed interval, the system first enters a delay judgment stage. Specifically, the system sets an observation window, which contains at least one subsequent AIS data point. Then, the system needs to determine whether the ground speed of all data points in the observation window continues to remain within the normal speed interval. For example, the system can configure the size of the observation window to be two consecutive AIS data points. Then, the system checks whether the ground speeds of the AIS data points at time T8 and T9 after the time T7 when the speed is detected to recover to normal are greater than or equal to the preset minimum speed threshold and less than or equal to the preset maximum speed threshold. Only when the speeds at time T8 and T9 both meet this condition, the system will generate an end signal indicating that the abnormal speed behavior starting at time T6 has ended. If at time T8, the speed meets the normal range, but at time T9, the speed falls back to the abnormal speed interval, the system will cancel the generation of the end signal and start monitoring again to avoid misjudgment. In other optional implementations, the size of the observation window can also be dynamically adjusted, for example, different window sizes can be set according to the type of the ship or the traffic density of the channel. In addition, the standard for "continuous retention" can also be more flexible, for example, a small amount (for example, 1) of short-term speed fluctuations within the observation window are allowed, as long as the overall trend is to remain within the normal speed interval, it can still be determined as "continuous retention".

[0148] Through the above scheme, the embodiment can effectively avoid misjudgment caused by AIS data fluctuation or ship speed fine-tuning, thereby improving the accuracy and reliability of the system in judging the end of the abnormal speed behavior.

[0149] In one specific implementation, based on the above embodiment, the method further comprises processing the boundary condition of the ship leaving the channel geographic fence. Specifically, when the processor judges that the geographic position of the previous data point is located within the channel geographic fence, and the geographic position of the current data point is located outside the geographic fence, the processor is configured to check whether there is an effective speed abnormal behavior associated with the target ship that has generated a warning signal but has not generated an end signal. If there is an effective speed abnormal behavior, the signal generation module generates an end signal to mark the complete life cycle of the abnormal behavior. In one specific implementation, the end signal can include an end reason field, the value of which is set to "leaving the channel" for subsequent event analysis and log recording.

[0150] In other optional implementations, in order to deal with the misjudgment caused by the drift of AIS data positioning near the boundary of the geographic fence, a buffer area can be set outside the geographic fence. When the geographic position of the previous data point is within the geofence, and the current data point is within the buffer zone, the same logic that generates the end signal is triggered.

[0151] With the above boundary processing scheme, the embodiment ensures that when the ship leaves the monitored water area, any active abnormal behavior event can be properly closed, thereby ensuring the logical integrity of the monitoring event life cycle and avoiding unnecessary continuous processing of ships that have left the monitoring range.

[0152] In a specific implementation, on the basis of the above embodiment, the waterway is pre-divided into multiple geographically non-overlapping sub-regions. This division is based on the actual physical conditions of the waterway and the navigation management needs. For example, the open water area at the entrance of the waterway can be divided into a sub-region, the narrow section within the waterway can be divided into another sub-region, and the turning area can be divided into a third sub-region.

[0153] For each sub-region, a set of independent speed thresholds and acceleration thresholds are maintained, which reflect the safety navigation requirements of the sub-region. For example, in open water, ships can travel at a higher speed, so a higher speed threshold and acceleration threshold can be set; while in narrow sections, ships need to travel at a lower speed and avoid sharp speed changes, so lower speed thresholds and acceleration thresholds are needed. These rule parameters associated with the sub-region are stored in the storage module and are associated with the corresponding geofence coordinates.

[0154] In other optional implementations, the sub-region division criteria can also be dynamically adjusted according to historical accident data, ship flow density, or hydro-meteorological conditions, etc. In addition, each sub-region can be associated with multiple sets of speed thresholds and acceleration thresholds, and can be switched according to different time periods or weather conditions to further improve the accuracy of identification.

[0155] With the above scheme, the embodiment can use different speed and acceleration thresholds for abnormal behavior identification according to the characteristics of different regions of the waterway, thereby more accurately reflecting the actual navigation risk of the ship and avoiding false positives and false negatives.

[0156] In a specific embodiment, on the basis of the above-mentioned embodiments, first, before judging whether the geographical positions of the current data point and the previous data point both fall within the geofence, the system determines the channel sub-region to which they both belong according to the latitude and longitude coordinates of the current data point and the previous data point. One method for determining the channel sub-region is to represent the geographical range of the channel sub-region as a polygon, and to use the algorithm for judging whether a point is within a polygon to judge whether the data point is located in a certain sub-region. If the current data point and the previous data point are both located within the polygon of the same sub-region, it is determined that they both belong to the sub-region.

[0157] Specifically, the system maintains a pre-configured mapping table of channel sub-regions and rule parameters, which is stored in the storage module 20. The mapping table takes the unique identifier of the channel sub-region as the key, and the corresponding value is the set of speed thresholds and the set of acceleration thresholds associated with the sub-region. Once the channel sub-region to which the current data point and the previous data point both belong is determined, the system retrieves the speed thresholds (including the minimum speed threshold and the maximum speed threshold) and the acceleration threshold associated with the sub-region from the mapping table. Then, the system will use these sub-region-specific thresholds for subsequent speed state judgment and speed change rate judgment, and then determine whether to generate a warning signal. In other optional embodiments, a machine learning-based classification model can also be used to automatically identify the channel sub-region in which the ship is located, which takes the latitude and longitude coordinates of the ship as input and outputs the channel sub-region to which the ship belongs.

[0158] Through the above scheme, the present embodiment can implement different speed and acceleration thresholds for different channel sub-regions, making the identification of abnormal behavior more refined and being able to adapt to the navigation characteristics of different segments, thereby improving the accuracy of the warning and reducing false positives.

[0159] In a specific embodiment, on the basis of the above-mentioned embodiment, the data acquisition module of the system is configured to specifically process AIS data streams from specific target vessels. First, the data acquisition module filters and groups all received AIS messages according to the vessel unique identification (MMSI, Maritime Mobile Service Identity) in the AIS message. Specifically, for each newly received AIS message, the data acquisition module extracts its MMSI and checks whether the MMSI already exists in a "live vessel session list" maintained in memory. If it exists, the message is added to the end of the AIS data stream corresponding to the MMSI. If it does not exist, a new AIS data stream is created, the MMSI is added to the "live vessel session list", and the message is taken as the first data point of the data stream. Then, the data acquisition module maintains a timestamp sequence for each AIS data stream to ensure that the AIS data points in each data stream are arranged in chronological order. Any out-of-order or duplicate data points are filtered or corrected. Next, the system can be configured with a "session timeout" mechanism. If an AIS data stream of a vessel has not received new data for a period of time (e.g., 10 minutes), it is considered that the vessel has left the monitoring area or the AIS device has failed, and the corresponding AIS data stream is marked as "inactive" and removed from the "live vessel session list" to release system resources. This mechanism avoids long-term occupation of system resources due to long-term stationary of vessels or interruption of AIS signals.

[0160] Through the above-mentioned scheme, the embodiment can ensure that the system processes continuous AIS data streams from the same vessel, thereby providing a reliable data basis for subsequent identification of abnormal speed behavior and avoiding misjudgment due to confusion of data from different vessels.

[0161] In a specific embodiment, on the basis of the above-mentioned embodiment, when the system determines that a warning signal needs to be generated, the data acquisition module encapsulates the extracted timestamp, geographic position (longitude and latitude), speed over ground, and type of abnormal speed interval in the current data point according to a predefined structured format. For example, the structured format can be JSON or XML. Specifically, an example of a JSON-formatted warning signal is as follows:

[0162] json

[0163] {

[0164] "timestamp": "2024-07-27T10:30:00Z",

[0165] "longitude": 121.4737,

[0166] "latitude": 31.2304,

[0167] "sog": 3.5,

[0168] "speed_exceedance_type": "Below minimum speed"

[0169] }

[0170] The JSON data set consists of several fields: timestamp (recording the time of the event, conforming to ISO 8601 standard); longitude and latitude (recording the ship's longitude and latitude coordinates, respectively, using the WGS 84 coordinate system); sog (recording the ship's speed over land, in knots); and speed_exceedance_type (recording the type of abnormal speed range, with possible values ​​of "below minimum speed" or "above maximum speed"). The signal generation module then sends this JSON-formatted data to a specified destination, such as the API interface of a maritime monitoring platform or a local log file. In other alternative implementations, the warning signal can also use other structured data formats, such as XML, or a custom binary format to further reduce the data volume. The type of the abnormal speed range can also be represented by a numeric code instead of a string, for example, 0 for "below minimum speed" and 1 for "above maximum speed".

[0171] Through the above scheme, this embodiment can provide structured early warning signals containing information such as time, location, speed and anomaly type, which facilitates maritime regulatory authorities to quickly understand the details of abnormal ship behavior and improve emergency response efficiency.

[0172] This invention provides a system for identifying abnormal speed behavior of ships entering and leaving ports, such as... Figure 5 As shown, the system includes a data acquisition module M100, a storage module M200, a processor M300, and a signal generation module M400, and is used to implement the identification system for abnormal ship speed behavior when entering and leaving port in any of the above embodiments.

[0173] The data acquisition module M100 is configured to obtain a series of time-ordered Automatic Identification System (AIS) data points containing the geographical position and the speed over ground of the target vessel. In one implementation, the data acquisition module M100 comprises a network interface capable of receiving AIS messages conforming to IEC 61162-1 and IEC 61993-2 standards. Inside the data acquisition module M100, a message parser is deployed to extract the fields of MMSI (Maritime Mobile Service Identity, a unique identifier of the vessel), time stamp, latitude and longitude coordinates, and speed over ground (SOG) from the received AIS messages. In other alternative implementations, the data acquisition module M100 can be directly connected to an AIS receiver through a serial port, or obtain data from a remote AIS base station through a wireless communication module (e.g. 4G / 5G module).

[0174] The storage module M200 is configured to store the geo-fences for defining the vessel routes on the map and the speed thresholds associated with the geo-fences. In one embodiment, the storage module M200 employs a PostgreSQL database with the PostGIS geographic information extension. The geo-fences are stored in the database in the form of polygons, each of which is composed of a series of latitude and longitude coordinate points. The speed thresholds are also stored in the database and associated with the corresponding geo-fences. These speed thresholds include the minimum and maximum speed thresholds for defining the normal speed range. In other alternative implementations, the storage module M200 can employ other types of databases (e.g. MySQL) or in-memory databases (e.g. Redis), and the geo-fences can be stored in other formats (e.g. GeoJSON).

[0175] The processor M300 is communicatively connected to the data acquisition module M100 and the storage module M200, and is configured to: select a current data point and a previous data point which are time-continuous in the AIS data point sequence, and determine whether the geographic positions of the current data point and the previous data point both fall within the geographic fence stored in the storage module M200; in response to the geographic positions of the current data point and the previous data point both falling within the geographic fence, determine, based on a threshold speed, whether the ground speed of the previous data point is in a normal speed interval and whether the ground speed of the current data point is in an abnormal speed interval. In one specific implementation, the processor M300 includes a geographic fence determination unit which uses a ray method to determine whether an AIS data point is located within a geographic fence. The processor M300 also includes a speed state determination unit which compares the ground speed of an AIS data point with the minimum speed threshold and the maximum speed threshold stored in the storage module M200, so as to determine whether the speed state of the data point is normal, too low or too high. In other optional embodiments, the processor M300 can use other types of geographic fence algorithms (such as a point-in-polygon algorithm) or use a hardware accelerator to improve the calculation speed.

[0176] The signal generation module M400 is communicatively connected to the processor M300, and is configured to: when the processor M300 determines that the ground speed of the previous data point is in the normal speed interval and the ground speed of the current data point is in the abnormal speed interval, generate a warning signal for indicating that the ship has a speed abnormal behavior. In one example, the signal generation module M400 includes an alarm information generator which can encapsulate the warning signal into an alarm sentence conforming to the NMEA 0183 protocol, and send the alarm sentence to a VTS (vessel traffic service) console through a serial port for display. The signal generation module M400 also includes an audible and visual alarm which will issue an audible and visual alarm when an abnormal behavior is detected. In other optional embodiments, the signal generation module M400 can send the warning signal to a remote monitoring center through an Ethernet interface, or send the alarm information to designated personnel through a short message gateway.

[0177] Through the above configuration, the data acquisition module M100 is responsible for continuously acquiring AIS data of the ship, the storage module M200 provides channel geographic information and speed rules, the processor M300 performs real-time analysis and determination based on these information, and the signal generation module M400 is responsible for issuing a warning when an abnormal behavior is detected. The various modules work together to achieve automatic identification and early warning of speed abnormal behaviors of ships entering and leaving the port. This system architecture can solve the problem that the prior art cannot effectively identify the sudden change process of the speed of the ship from normal to abnormal, thereby achieving earlier and more accurate risk warning.

[0178] By the above scheme, the embodiment can monitor the ship speed in real time, and timely issue a warning when detecting that the speed abnormally mutates, thereby effectively improving the navigation safety guarantee capability of the port channel. Compared with the prior art, the embodiment has the beneficial effects that the "mutation process" of the speed can be identified instead of only the "abnormal state", so that the warning timing can be significantly advanced, and sufficient time is provided for timely intervention measures.

[0179] In a specific embodiment, on the basis of the above embodiment, the processor M300 is further configured to, after the signal generation module M400 generates the warning signal, continuously acquire subsequent AIS data points of the ground speed from the data acquisition module M100, and repeatedly execute the logic of the speed state judgment. Each time a new AIS data point is received, the processor M300 judges whether the ground speed of the data point is still in the abnormal speed interval based on the speed threshold value stored in the storage module M200. The signal generation module M400 is configured to generate an end signal indicating the end of the speed abnormal behavior only when the processor M300 detects that the ground speed recovers from the abnormal speed interval to the normal speed interval.

[0180] In a specific example, it is assumed that the processor M300 detects that the ship has a "sudden speed reduction" behavior at time T1, and instructs the signal generation module M400 to generate a corresponding warning signal. Thereafter, the processor M300 continues to receive and analyze subsequent AIS data points. If the ground speed of the received AIS data point is still lower than the preset minimum speed threshold at T2, the processor M300 will not take any action. Until T3, the ground speed of the AIS data point received by the processor M300 is higher than the minimum speed threshold, indicating that the ship has recovered to normal speed. At this time, the processor M300 instructs the signal generation module M400 to generate an end signal indicating the end of the speed abnormal behavior.

[0181] In other optional embodiments, the processor M300 can be configured to, after detecting that the ground speed recovers to the normal speed interval, not immediately instruct the signal generation module M400 to generate the end signal, but to increase a delay confirmation mechanism. For example, only after continuously receiving N (N is an integer greater than 1, such as 2, 3 or 5) AIS data points of the ground speed higher than the minimum speed threshold, it is finally confirmed that the speed abnormal behavior has ended, and the corresponding end signal is generated. The delay confirmation mechanism can be realized by a counter or a timer. The counter records the number of continuous normal speed data points, and when the counter reaches the preset threshold N, the end signal is generated. The timer records the time when the speed recovers to normal, and only when the time exceeds the preset delay time, the end signal is generated. The number N of the above delay confirmation can be configured through the storage module.

[0182] Through the above scheme, the embodiment can avoid misjudgment caused by AIS data fluctuation or instantaneous adjustment of ship speed, and improve the accuracy and reliability of the end judgment of abnormal speed behavior.

[0183] In a specific embodiment, on the basis of the above embodiment, the storage module M200 not only stores the channel geographic fence and the speed threshold, but also stores the acceleration threshold associated with the geographic fence. For example, in the PostgreSQL database, an "acceleration_threshold" field can be added to each channel record. The data type of this field can be a floating point number with units of knots per minute, which represents the maximum acceleration allowed for the channel. When the processor M300 judges whether the ship speed is abnormal, it first reads the acceleration threshold of the channel from the storage module M200, and calculates the speed change rate according to the current AIS data point and the previous AIS data point provided by the data acquisition module M100. The calculation method can be: (current speed - previous speed) / (current timestamp - previous timestamp). After obtaining the speed change rate, the processor M300 judges whether its absolute value is greater than the acceleration threshold obtained from the storage module M200. Only when the absolute value of the speed change rate is greater than the threshold, and the condition of speed state transition is met, the signal generation module M400 generates a warning signal. In some other optional embodiments, the acceleration threshold can be stored in a separate configuration file or dynamically obtained from an external system through an API interface, rather than directly stored in the database.

[0184] Through the above scheme, the embodiment can effectively distinguish between "sudden" speed change and "slow" speed change of the ship, thereby improving the accuracy of the warning and reducing false positives.

[0185] In a specific embodiment, on the basis of the above-mentioned embodiments, the geo-fence stored in the storage module M200 is divided into a plurality of channel sub-regions. Each channel sub-region is associated with a set of independent speed threshold values, including but not limited to the minimum speed threshold value, the maximum speed threshold value. The division of the geo-fence can be based on the physical characteristics of the channel (for example, channel width, water depth, curvature) or traffic management strategies (for example, traffic control area, dangerous goods transportation area). The boundary of each channel sub-region is defined by a set of latitude and longitude coordinate points and stored in the storage module M200. The processor M300 first determines the channel sub-region to which the current data point and the previous data point belong according to the geographical positions of the current data point and the previous data point before determining whether the geographical positions of the current data point and the previous data point are both within the geo-fence. Specifically, the processor M300 uses the point-in-polygon algorithm to determine whether the current data point and the previous data point are both within the geo-fence of a specific channel sub-region. If the result is yes, the processor M300 calls the set of speed threshold values associated with the channel sub-region for subsequent speed state judgment. If the two data points belong to different channel sub-regions, the system can use a more refined interpolation algorithm to calculate a weighted average speed threshold value according to the distance weight, or directly ignore this judgment and wait until both data points enter the same sub-region before making a judgment.

[0186] Through the above scheme, the present embodiment can set different speed threshold values for different channel sub-regions, so that the identification of abnormal behavior of ships entering and leaving the port is more in line with the actual navigation environment, and the accuracy and reliability of the warning are improved.

[0187] The above is a further detailed description of the present application in combination with specific preferred embodiments, and the specific implementation of the present application cannot be limited to these descriptions. For ordinary skilled persons in the technical field to which the present application belongs, without departing from the concept of the present application, a number of simple deductions or substitutions can be made, which should be considered as falling within the scope of protection of the present application.

Claims

1. A method of identifying abnormal behavior of a ship's approach and departure speed from a port, characterized by, The method comprises: acquiring a series of time-sequenced automatic identification system (AIS) data points containing geographic positions and ground speeds of a target ship; based on a pre-stored geographic fence used to define a ship route on a map and a ground speed threshold value associated with the geographic fence; selecting a current data point and a previous data point in the series of AIS data points in time sequence, and determining whether the geographic positions of the current data point and the previous data point both fall within the geographic fence; in response to the geographic positions of the current data point and the previous data point both falling within the geographic fence, determining, based on the ground speed threshold value, whether the ground speed of the previous data point is within a normal speed range and the ground speed of the current data point is within an abnormal speed range; when the ground speed of the previous data point is within the normal speed range and the ground speed of the current data point is within the abnormal speed range, generating a warning signal indicating that the ship has an abnormal speed behavior.

2. The method of claim 1, wherein, The abnormal speed range is lower than a preset minimum speed threshold value.

3. The method of claim 1, wherein, The abnormal speed range is higher than a preset maximum speed threshold value.

4. The method of claim 1, wherein, The method further comprises: after generating the warning signal, continuously acquiring and determining the ground speed of a subsequent AIS data point, and when it is detected that the ground speed has returned to the normal speed range from the abnormal speed range, generating an end signal indicating that the abnormal speed behavior has ended.

5. The method of claim 1, wherein, The method further comprises: after the step of determining whether the geographic positions of the current data point and the previous data point both fall within the geographic fence, when it is determined that the geographic position of the previous data point is outside the geographic fence and the geographic position of the current data point is within the geographic fence, not performing the step of generating the warning signal.

6. The method of claim 1, wherein, The method further comprises: before the step of generating the warning signal, the method further comprises: based on a pre-stored acceleration threshold value associated with the geographic fence, calculating a speed change rate between the current data point and the previous data point, and determining whether the absolute value of the speed change rate is greater than the acceleration threshold value; the warning signal is generated when the absolute value of the speed change rate is greater than the acceleration threshold value.

7. The method of claim 6, wherein, The step of calculating the speed change rate comprises: extracting respective time stamps from the current data point and the previous data point; calculating a time difference based on the time stamps; dividing the difference between the ground speeds of the current data point and the previous data point by the time difference to obtain the speed change rate.

8. The method of claim 1, wherein, The pre-stored geographic fence is divided into a plurality of route sub-regions, and a set of ground speed threshold values is independently associated with each route sub-region.

9. The method of claim 8, wherein, The method further comprises: before the step of determining whether the geographic positions of the current data point and the previous data point both fall within the geographic fence, first determining the route sub-region to which the current data point and the previous data point both belong according to their geographic positions, and calling the set of ground speed threshold values associated with the route sub-region for subsequent determination.

10. A system for identifying abnormal behavior of a ship's approach and departure speed from a port, characterized by The method comprises: a data collection module configured to acquire a series of time-sequenced automatic identification system (AIS) data points containing geographic position and ground speed of a target ship; a storage module configured to store a geo-fence for defining a ship route on a map and a speed threshold associated with the geo-fence; a processor communicatively connected to the data collection module and the storage module, the processor being configured to: select a current data point and a previous data point that are time-continuous in the series of AIS data points, and determine whether the geographic positions of the current data point and the previous data point both fall within the geo-fence stored in the storage module; in response to the geographic positions of the current data point and the previous data point both falling within the geo-fence, determine, based on the speed threshold, whether the ground speed of the previous data point is within a normal speed range and whether the ground speed of the current data point is within an abnormal speed range; a signal generation module communicatively connected to the processor, the signal generation module being configured to: generate a warning signal for indicating that the ship has an abnormal speed behavior when the processor determines that the ground speed of the previous data point is within the normal speed range and the ground speed of the current data point is within the abnormal speed range.

Images