AIS track data processing method for real-time monitoring of ship situation

Abstract

The invention discloses an AIS track data processing method for real-time monitoring of a ship situation. The method comprises the following steps: step 1, receiving and analyzing AIS data; step 2, AIS abnormal data cleaning; step 3, AIS track area attribute determination; 4, judging a maneuvering mode; step 5, carrying out self-adaptive statistics on the life cycle; step 6, pre-pushing; step 7, carrying out self-adaptive statistics on a pre-pushing period; step 8, ending pre-pushing; and step 9, updating and storing AIS static information. According to the method, data management of AIS tracks in a ship traffic management system is realized, abnormal data such as AIS jumping and fallback are eliminated, information such as historical updating time intervals and navigational speed changes of the AIS is comprehensively considered, the pre-pushing cycle and the life cycle of the AIS are subjected to self-adaptive statistics, and meanwhile, the pre-pushing cycle and the life cycle of the AIS are calculated in combination with geographic element information of a sea graph layer. AIS track data processing is carried out according to regions, and the effects of AIS track smoothness and high prediction precision are well achieved.

Description

Technical Field

[0001] The invention relates to the technical field of real-time monitoring of ships, in particular to an AIS track data processing method for real-time monitoring of ship situations. Background Art

[0002] AIS (Automatic Identification System) is an auxiliary navigation device that is mandatory to be installed on ships according to the International Maritime Organization. It has the advantage of working around the clock and in all weather conditions. AIS uses open broadcast technology, and the broadcast data includes dynamic information such as the ship's position, heading, speed, and static information such as the ship's name, ship type, and other safety information. However, due to the open data transmission characteristics of AIS itself, the AIS base station forwarding packet loss or data reception frame loss, as well as various factors that affect ships during navigation, such as poor signals and interference, AIS data is often inaccurate or adjacent AIS information has a large time span or is even lost.

[0003] The ship traffic management system accesses AIS data and monitors the ship traffic in ports, inland rivers and coastal areas in real time at the traffic control terminal to improve the safety of ship navigation. The ship traffic management system monitors the ship's situation in real time, and needs to keep the ship's movement trajectory and situation as consistent as possible with the actual situation to ensure the smoothness of the ship's trajectory. The AIS device periodically sends ship movement information and automatically adjusts the information sending time according to the ship's movement status. Therefore, the AIS data update time interval is not fixed. In addition to the above reasons, the AIS trajectory will have target freezes, jumps, or abnormal convex points. The track is discontinuous, which is not conducive to real-time monitoring of ship status, thereby avoiding collisions, predicting risks, and avoiding water traffic accidents. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an AIS track data processing method for real-time monitoring of ship status in view of the deficiencies of the above-mentioned prior art. The AIS track data processing method for real-time monitoring of ship status eliminates abnormal data such as AIS jumps, fallbacks, and redundancies, and comprehensively considers information such as the historical update time interval and speed changes of AIS, adaptively calculates the prediction period and life cycle of AIS, and at the same time combines the geographical element information of the nautical chart layer to process the AIS track data by region, thereby achieving a smooth AIS track and high prediction accuracy.

[0005] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0006] A method for processing AIS track data for real-time monitoring of ship status comprises the following steps.

[0007] Step 1. Receive and parse AIS data: The parsed AIS data includes AIS dynamic information and AIS static information; wherein the AIS dynamic information includes latitude and longitude, time, heading and speed.

[0008] Step 2: Cleaning up abnormal AIS data: Remove abnormal update data from the parsed AIS data.

[0009] Step 3, AIS track area attribute determination: According to the longitude and latitude in the AIS dynamic information obtained by step 1, and the regional attribute information of the geographic element grid, the regional attribute information of the target ship at the current position is obtained; wherein the regional attributes include waters, land, bridges, anchorages and berths.

[0010] Step 4: Determine the maneuvering mode: Based on the time and speed information in the AIS dynamic information, determine the maneuvering mode of the target ship at the current moment; wherein the maneuvering mode includes uniform speed, acceleration and deceleration.

[0011] Step 5: Adaptive statistics of life cycle: According to the maneuvering mode obtained in step 4 and the AIS track area attributes in step 2, adaptive statistics of the target ship life cycle are performed, specifically:

[0012] A. When the maneuvering mode is uniform speed or acceleration, and the AIS track area attribute is water area, the target ship life cycle t Life The adaptive statistical method comprises the following steps:

[0013] Step 5A-1: Count the maximum update time interval t max :Collect n historical dynamic update trajectory points, count the n historical update cycle intervals Δt, and record the maximum update cycle interval Δt as t max .

[0014] Step 5A-2: Calculate t Life , the specific calculation formula is:

[0015] t Life =min(max(μ*t max ,t Lifemin ),t Lifemax )

[0016] Where μ is the adaptive statistical coefficient.

[0017] t Lifemin is the lower threshold of the life cycle, t Lifemax The upper threshold of the life cycle.

[0018] B. When any of the following conditions A or B occurs, the adaptive statistical coefficient μ needs to be increased; where:

[0019] Condition A: When the AIS track area attribute is anchorage or berth.

[0020] Condition B: When the maneuvering mode is deceleration.

[0021] At this time, the target ship life cycle t Life The calculation formula is:

[0022] t Life =min(max(μ'*t max ,t Lifemin ),t Lifemax )

[0023] in:

[0024] μ'>μ

[0025] Step 6: Pre-push: When the maneuvering mode is uniform speed, after the current AIS dynamic update, the AIS dynamic information of the target ship is pre-pushed once every collection time period ΔT; wherein, the collection time period ΔT is consistent with the radar data update cycle.

[0026] Step 7: Adaptive statistical prediction cycle: Predict the target ship for prediction cycle t Fore The adaptive statistics of , the specific calculation formula is:

[0027]

[0028] In the formula, is the average value within n AIS update cycles,

[0029] t Poremin is the lower limit threshold of the prediction period, t Poremax The upper threshold of the pre-push period.

[0030] Step 8, end pre-pushing: For the target ship of pre-pushing, if any of the following conditions occurs during the pre-pushing process, the pre-pushing will be ended, specifically:

[0031] Condition 1: The pre-pushing time reaches the pre-pushing period t calculated in step 7 Fore .

[0032] Condition 2: The geographical element area attribute of the target location predicted by the AIS track is located on land.

[0033] Step 9: AIS static information update and storage.

[0034] In step 2, the abnormal update data includes AIS jump abnormal data, AIS fallback abnormal data and AIS repeated update data.

[0035] In step 2, the method for removing AIS jump abnormal data includes the following steps:

[0036] Step 2A-1: Set the AIS jump abnormality adaptive threshold D m , the specific calculation formula is:

[0037] D m =ζ*max(V0,V1)*(T1-T0)

[0038] Where ζ is the AIS position jump anomaly coefficient and is the set value.

[0039] V0 and T0 are the speed value and update time value of the AIS target in the data of the last dynamic update time.

[0040] V1 and T1 are the speed value and update time value of the AIS target in the current dynamic update data.

[0041] Step 2A-2: Calculate the distance D between the last dynamic update time and the current dynamic update time of the AIS target is .

[0042] Step 2A-3, AIS jump abnormality judgment: D is With D m For comparison, when D is >D m When the AIS dynamic update data is abnormal, it is judged that the current AIS dynamic update data is abnormal and is removed.

[0043] Step 2A-4, according to the maneuvering mode obtained in step 4, automatically adjust the AIS position jump anomaly coefficient ζ, specifically: A. When the maneuvering mode is uniform speed or deceleration, taking into account the influence of wind speed and water flow, increase fault tolerance, and ζ takes the conventional setting value;

[0044] B. When the maneuvering mode is acceleration, the AIS position jump anomaly coefficient ζ adaptively adjusts the acceleration a of the target ship. Assuming the adaptively adjusted AIS position jump anomaly coefficient is ξ', the specific calculation formula of ξ' is:

[0045] ξ'>ξ,ξ'=(1+a)*ξ.

[0046] In step 2, the method for removing abnormal data from AIS rollback includes the following steps:

[0047] Step 2B-1, calculate the turning degree C m :Set N AIS dynamic update trajectory points, collect and calculate the turning degree C of the AIS target and the historical update trajectory points m .

[0048] Step 2B-2, AIS fallback abnormality judgment: turn the steering angle C m The steering threshold Chold For comparison; when C m >C hold , it is judged as AIS rollback abnormality and the AIS update data is discarded.

[0049] In step 2, the method for eliminating AIS duplicate update data is as follows: set the speed value of the AIS target's last dynamic update data to V0, the update time to T0, and the longitude to L on0 , latitude is L at0 In this dynamic update, the speed value is V1, the update time is T1, and the longitude is L on1 , latitude is L at1 ; When (T1-T0)<T d 、V1>0 and L on1 =L on0 ,L at1 =L at0 , the AIS update data is considered to be repeated, and the dynamic update time data is removed.

[0050] Step 3: A method for determining the attributes of the AIS track area includes the following steps:

[0051] Step 3-1, divide into large grids: divide the monitored sea area into large grids of size a*a.

[0052] Step 3-2, export layer geographic elements: download all layer geographic elements with regional attribute information within the monitored sea area from the nautical chart.

[0053] Step 3-3, refine the grid once: divide each large grid that has an intersection relationship with the geographic elements of the layer exported in step 3-2 into four equal parts and refine them evenly to form a small refined grid once.

[0054] Step 3-4, secondary refined grid: Each primary refined small grid that has an intersection relationship with the layer geographic elements exported in step 3-2 is divided into four equal parts and refined evenly to form secondary refined small grids.

[0055] Step 3-5, m-times refined grid: repeat step 3-4, and refine the secondary refined small grid that has an intersection relationship with the layer geographic elements exported in step 3-2 m times to form m-times refined small grids; wherein the side lengths of the m-times refined small grids are all greater than the set size b; wherein a>b.

[0056] Step 3-6, geographic element grid regional attribute storage: the grid codes and regional attributes of all large grids, as well as the regional attributes of the refined small grids are serialized and saved in files.

[0057] Step 3-7, AIS track area attribute determination: When the AIS track is updated and predicted, the large grid code to which the target belongs is located according to the latitude and longitude information in the AIS dynamic information, and the area attribute information to which the target position finally belongs is queried in real time.

[0058] a=2000 meters, b=10 meters.

[0059] In step 5, μ'=2μ.

[0060] In step 9, the AIS static information updating and storage method includes the following steps:

[0061] Step 9-1, load local file: The local file stores AIS historical static data, including MMSI, English name, call sign, IMO, length and width information. After loading, AIS static data is saved to cache data with MMSI as the primary key.

[0062] Step 9-2, AIS static information judgment: The AIS static information parsed by the AIS in step 1 is judged to be null and valid respectively; when the parsed static information is null, the corresponding static data is queried in the cache data of step 9-1 and used; when the parsed static information is not null and is valid, the latest parsed static information is used, and at the same time, the static information of the local cache data in step 9-1 is updated to ensure the validity and real-time nature of the static information of the local cache data.

[0063] Step 9-3, update local files regularly: set the time interval T save , update the local cache static data updated in step 9-1 to the local file.

[0064] The present invention has the following beneficial effects:

[0065] The present invention uses different processing mechanisms to determine and eliminate erroneous abnormal data such as AIS jumps, fallbacks, and redundancies caused by problems such as AIS positioning equipment, base station forwarding, and data transmission, and comprehensively considers information such as the historical update time interval and speed change of AIS, adaptively calculates the prediction cycle and life cycle of AIS, and combines the geographical element information of the nautical chart layer to process AIS track data in different regions, which better meets the requirements of actual application scenarios for real-time trajectory prediction. Adopting an adaptive statistical prediction cycle mechanism, the prediction time interval is consistent with the radar sampling cycle, which largely ensures the consistency of the AIS trajectory and the radar echo, and timely and effectively predicts the navigation status of the ship, which helps to avoid water traffic accidents, enhance ship supervision capabilities, and ensure water traffic safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Figure 1The flowchart of the AIS track data processing method for real-time monitoring of ship status of the present invention is shown.

[0067] Figure 2 AIS fallback anomaly diagram is shown.

[0068] Figure 3 A schematic diagram of the AIS pre-launch cycle and life cycle is shown.

[0069] Figure 4 A schematic diagram showing the end of the AIS life cycle after long periods of non-update is shown.

[0070] Figure 5 A schematic diagram of the AIS adaptive statistical prediction cycle and life cycle history update interval is shown.

[0071] Figure 6 Shows a schematic diagram of the gridding of geographic features in the chart layer.

[0072] Figure 7 A schematic diagram of local storage of AIS static information is shown. DETAILED DESCRIPTION

[0073] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific preferred embodiments.

[0074] like Figure 1 As shown, a method for processing AIS track data for real-time monitoring of ship status includes the following steps.

[0075] Step 1. Receive and parse AIS data: The parsed AIS data includes AIS dynamic information and AIS static information. AIS dynamic information includes latitude and longitude, time, heading, speed and ship heading, etc. AIS static information includes MMSI, ship name, ship type, call sign, IMO, ship length and ship width, etc.

[0076] Step 2: Clean up abnormal AIS data: remove abnormal update data from the parsed AIS data.

[0077] The above abnormal update data includes AIS jump abnormal data, AIS fallback abnormal data and AIS repeated update data, etc.

[0078] Set the speed value of the AIS target in the last dynamic update data to V0, the update time to T0, and the longitude to L on0 , latitude is L at0 , the longitude and latitude are converted into the position information in the rectangular coordinate system as d x0 ,d y0 In this dynamic update, the speed value is V1, the update time is T1, and the longitude is L on1 , latitude is L at1, the longitude and latitude are converted into the position information in the rectangular coordinate system as d x1 ,d y1 The reference point longitude used to convert the geographic coordinate system into the rectangular coordinate system is CenterLon, and the reference point latitude is CenterLat.

[0079] The above-mentioned method for eliminating abnormal AIS jump data includes the following steps.

[0080] Step 2A-1: Set the AIS jump abnormality adaptive threshold D m , the specific calculation formula is:

[0081] D m =ζ*max(V0,V1)*(T1-T0)

[0082] Where V0 and T0 are the speed value and update time value of the AIS target in the last dynamic update data.

[0083] V1 and T1 are the speed value and update time value of the AIS target in the current dynamic update data.

[0084] ζ is the AIS position jump anomaly coefficient, which is the set value.

[0085] Step 2A-2: Calculate the distance D between the last dynamic update time and the current dynamic update time of the AIS target is , as shown below:

[0086]

[0087] Step 2A-3, AIS jump abnormality judgment: D is With D m For comparison, when D is >D m When the AIS dynamic update data is abnormal, it is judged that the current AIS dynamic update data is abnormal and is removed.

[0088] Step 2A-4, according to the maneuvering mode obtained in step 4, automatically adjust the AIS position jump anomaly coefficient ζ, specifically: A. When the maneuvering mode is uniform speed or deceleration, taking into account the influence of wind speed and water flow, increase the fault tolerance, and ζ is preferably 1.5; B. When the maneuvering mode is acceleration, the AIS position jump anomaly coefficient ζ needs to be increased, and the greater the acceleration, the greater ζ; at this time, the AIS jump anomaly adaptive threshold D m The calculation formula is:

[0089] D m =ζ'*max(V0,V1)*(T1-T0)

[0090] in:

[0091]

[0092] The above-mentioned method for removing abnormal AIS rollback data includes the following steps.

[0093] Step 2B-1, calculate the turning degree C m :Set N AIS dynamic update trajectory points, collect and calculate the turning degree C of the AIS target and the historical update trajectory points m .

[0094] like Figure 2 As shown, N track points are set, including historical update track points and current update track points. In this embodiment, N=3 is optimized. Figure 2 As shown, the vector P is constructed by the position information between the two adjacent points. AB [0], P AB [1], the angle θ1 between two vectors is calculated according to the vector angle calculation formula, then

[0095] θ1=arccos((P AB [0]·P AB [1]) / (P AB [0]*P AB [1])

[0096] C m =θ1 / π*180

[0097] Step 2B-2, AIS fallback abnormality judgment: turn the steering angle C m The steering threshold C hold For comparison; when C m >C hold When the turning degree threshold is set to C, it is judged that the AIS rollback is abnormal, and the AIS update data is discarded. hold =160°.

[0098] The system supports simultaneous access to multiple AIS data sources, such as the Aviation Insurance Center, multiple AIS base stations, etc. Therefore, at the same time or in a short time T d Multiple AIS messages with the same data information may be received at the same time. To improve processing performance, d The received speed is greater than a certain speed threshold V d , and when the AIS dynamic message with unchanged longitude and latitude information is received, it is determined that duplicate AIS dynamic update data has been received and deduplication processing is performed at the same time.

[0099] The method for eliminating the above AIS duplicate update data is as follows: set the speed value of the AIS target's last dynamic update time data to V0, the update time to T0, and the longitude to L on0, latitude is L at0 In this dynamic update, the speed value is V1, the update time is T1, and the longitude is L on1 , latitude is L at1 ; When (T1-T0)<T d 、V1>0 and L on1 =L on0 ,L at1 =L at0 , the AIS update data is considered to be repeated, and the dynamic update time data is removed.

[0100] Step 3, AIS track area attribute determination: According to the longitude and latitude in the AIS dynamic information obtained by step 1, and the regional attribute information of the geographic element grid, the regional attribute information of the target ship at the current position is obtained; wherein the regional attributes include waters, land, bridges, anchorages and berths, etc.

[0101] The above-mentioned method for determining the attributes of the AIS track area includes the following steps.

[0102] Step 3-1, divide into large grids: divide the monitored sea area into large grids of size a*a. In this embodiment, preferably a=2000 meters.

[0103] Step 3-2, export layer geographic elements: download all layer geographic elements with regional attribute information within the monitored sea area from the nautical chart.

[0104] Step 3-3, refine the grid once: divide each large grid that has an intersection relationship with the geographic elements of the layer exported in step 3-2 into four equal parts and refine them evenly to form a small refined grid once.

[0105] There are three types of relationships between large grids and layer geographic elements: the large grids intersect with layer geographic elements, the large grids are completely outside all layer geographic elements, and the large grids are completely inside a certain layer geographic element.

[0106] When the large grid is completely outside all the layer geographic elements or the large grid is completely inside a layer geographic element, its large grid area attributes are saved; when the large grid has an intersection relationship with the layer geographic elements, the large grid needs to be further refined, as shown below: Figure 6 shown.

[0107] Step 3-4, secondary refined grid: Each primary refined small grid that has an intersection relationship with the layer geographic elements exported in step 3-2 is divided into four equal parts and refined evenly to form secondary refined small grids.

[0108] Step 3-5, m-fold refined grid: repeat step 3-4, and refine the secondary refined small grid that has an intersection relationship with the layer geographic elements exported in step 3-2 m times to form m-fold refined small grids; wherein the side lengths of the m-fold refined small grids are all greater than the set size b; wherein a>b. In this embodiment, preferably b=10 meters.

[0109] When the final small grid still intersects with the layer's geographic features, it will no longer be refined, and the default attribute of this small grid area is water area.

[0110] Step 3-6, geographic element grid regional attribute storage: the grid codes and regional attributes of all large grids, as well as the regional attributes of the refined small grids are serialized and saved in files.

[0111] Step 3-7, AIS track area attribute determination: When the AIS track is updated and predicted, the large grid code to which the target belongs is located according to the latitude and longitude information in the AIS dynamic information, and the area attribute information to which the target position finally belongs is queried in real time.

[0112] When an AIS track appears for the first time, if the geographical element area attribute of its target location is located on land, it is judged as abnormal data and will not be processed, so as to avoid the appearance of AIS targets located on land at the traffic control terminal.

[0113] Step 4: Determine the maneuvering mode: Based on the time and speed information in the AIS dynamic information, determine the maneuvering mode of the target ship at the current moment; wherein the maneuvering mode includes uniform speed, acceleration and deceleration.

[0114] According to the AIS international standard ITU-RM.1371-4, the reporting interval of ship mobile equipment decreases as the speed increases. The lower the speed, the longer the reporting interval.

[0115] Set N update trajectory points, including historical update trajectory points and current real-time update trajectory points, and N=3 in this article.

[0116] The speed information of the target’s N trajectory points is (V n ,T n ), where V n is the speed information of the Nth track point, T n is the timestamp information of the Nth trajectory point, then

[0117] a n-1 =(V n-1 -V n ) / (T n-1 -T n )

[0118] Among them, the value range of n is [1, N], when n = 1, it is the current update trajectory point, a n-1Update the acceleration of the n-1th trajectory point.

[0119] N trajectory update points can get N-1 acceleration values, and the N-1 acceleration values ​​are compared with the acceleration threshold a one by one. hold Compare and record the acceleration value that is less than the acceleration threshold a hold The total number N num , if N num =(N-1), then the deceleration maneuver model is satisfied.

[0120] Step 5: Adaptive statistics of life cycle: According to the maneuvering mode obtained in step 4 and the AIS track area attributes in step 2, adaptive statistics of the target ship life cycle are performed, specifically:

[0121] A. When the maneuvering mode is uniform speed or acceleration, and the AIS track area attribute is water area, the target ship life cycle t Life The adaptive statistical method comprises the following steps.

[0122] Step 5A-1: Count the maximum update time interval t max

[0123] First, collect n historical dynamic update trajectory points, and count the historical n update cycle intervals Δt, which are recorded as: Δt1, Δt2, Δt3, ..., Δt n , such as Figure 5 shown.

[0124] Next, the maximum update cycle interval Δt is recorded as t max , that is:

[0125] t max =max(Δt1,Δt2,Δt3,...,Δt n )

[0126] Step 5A-2: Calculate t Life , the specific calculation formula is:

[0127] t Life =min(max(μ*t max ,t Lifemin ),t Lifemax )

[0128] Wherein, μ is an adaptive statistical coefficient, preferably μ=1.0.

[0129] t Lifemin is the lower threshold of the life cycle, t Lifemax The upper threshold of the life cycle.

[0130] B. When any of the following conditions A or B occurs, the adaptive statistical coefficient μ needs to be increased; where:

[0131] Condition A: When the AIS track area attribute is an anchorage or berth, increase μ to avoid misjudgment of AIS shutdown or loss due to a longer AIS reporting interval when the AIS is located at an anchorage or berth.

[0132] Condition B: When the maneuvering mode is deceleration, the reporting interval is extended. At this time, in order to avoid the life cycle calculated by adaptive statistics being too short and misjudged as target loss, an automatic life cycle extension mechanism is adopted.

[0133] At this time, the target ship life cycle t Life The calculation formula is:

[0134] t Life =min(max(μ'*t max ,t Lifemin ),t Lifemax )

[0135] Among them, μ'>μ, in this embodiment, μ'=2μ is preferred.

[0136] The entire process cycle of an AIS target from its first resolution to its loss is called the AIS lifetime.

[0137] When the AIS information is updated, the AIS life cycle t Life Recalculate the timing, so in the AIS life cycle t Life As long as the AIS is updated in time, the AIS life span will continue. Figure 3 Only when AIS is not updated for a long time and the time from the last update exceeds the life cycle t Life After that, it is considered that the AIS is turned off or the device is abnormal, and the target is lost. Figure 4 shown.

[0138] Step 6: Pre-launch

[0139] The AIS data update interval is irregular, and the real-time information of the ship is updated slowly. In order to make the ship trajectory smooth, when the maneuvering mode is uniform speed, after the current AIS dynamic update, the AIS dynamic information of the target ship is predicted once every collection time period ΔT; among which, the collection time period ΔT is consistent with the radar data update cycle, and ΔT=0.25s is preferably selected.

[0140] Step 7: Adaptive statistical prediction cycle: Predict the target ship for prediction cycle t Fore The adaptive statistics of , the specific calculation formula is:

[0141]

[0142] in:

[0143]

[0144] In the formula, It is the average value within n AIS update cycles.

[0145] t Poremin is the lower limit threshold of the prediction period, t Poremax The upper threshold of the pre-push period.

[0146] The above AIS pre-launch period is shorter than the life cycle.

[0147] Step 8, end pre-pushing: For the target ship of pre-pushing, if any of the following conditions occurs during the pre-pushing process, the pre-pushing will be ended, specifically:

[0148] Condition 1: As the prediction time increases, the prediction accuracy gradually decreases. When the prediction time reaches the prediction period t calculated in step 7, Fore , then stop pre-pushing.

[0149] Condition 2: The target location of the AIS track prediction belongs to a geographical element whose regional attribute is located on land. At this time, the prediction is stopped to avoid the prediction landing.

[0150] Step 9: AIS static information update and storage.

[0151] like Figure 7 As shown, the AIS static information updating and storage method includes the following steps.

[0152] Step 9-1, load local file: AIS historical static data is stored in the local file, including MMSI, English name, call sign, IMO, length and width information. After loading, AIS static data is saved to cache data with MMSI as the primary key.

[0153] Step 9-2, AIS static information judgment: The AIS static information parsed by the AIS in step 1 is judged to be null and valid respectively; when the parsed static information is null, the corresponding static data is queried in the cache data of step 9-1 and used; when the parsed static information is not null and is valid, the latest parsed static information is used, and at the same time, the static information of the local cache data in step 9-1 is updated to ensure the validity and real-time nature of the static information of the local cache data.

[0154] Step 9-3, update local files regularly: set the time interval T save , update the local cache static data updated in step 9-1 to the local file. In this embodiment, preferably T save =1 hour.

[0155] The present invention locally stores and manages the static information parsed by AIS, including the English name, call sign, IMO, ship length, ship width and other information, so that when the AIS target appears again, comprehensive static information can be obtained in time, avoiding incomplete information caused by untimely and incomplete static data transmission.

[0156] In summary, the present invention adopts different processing mechanisms to determine and eliminate erroneous abnormal data such as AIS jump, fallback, redundancy, etc. caused by problems such as AIS positioning equipment, base station forwarding, and data transmission, and comprehensively considers information such as the historical update time interval and speed change of AIS, adaptively calculates the prediction cycle and life cycle of AIS, and combines the geographic element information of the nautical chart layer to process AIS track data by region, thereby achieving the effect of smooth AIS trajectory and high prediction accuracy, better meeting the requirements of actual application scenarios for real-time trajectory prediction, and ensuring water transportation safety.

[0157] The preferred embodiments of the present invention are described in detail above; however, the present invention is not limited to the specific details in the above embodiments. Within the technical concept of the present invention, various equivalent transformations can be made to the technical solutions of the present invention, and these equivalent transformations all belong to the protection scope of the present invention.

Claims

1. A method for processing AIS track data for real-time monitoring of ship status, characterized in that: The steps include: Step 1, receiving and parsing AIS data: the parsed AIS data includes AIS dynamic information and AIS static information; wherein the AIS dynamic information includes latitude and longitude, time, heading and speed; Step 2: Cleaning up abnormal AIS data: removing abnormal update data from the parsed AIS data; Step 3, AIS track area attribute determination: according to the latitude and longitude in the AIS dynamic information obtained by step 1, and the regional attribute information of the geographic element grid, the regional attribute information of the target ship at the current position is obtained; wherein the regional attributes include waters, land, bridges, anchorages and berths; Step 4, determine the maneuvering mode: based on the time and speed information in the AIS dynamic information, determine the maneuvering mode of the target ship at the current moment; wherein the maneuvering mode includes uniform speed, acceleration and deceleration; Step 5: Adaptive statistics of life cycle: According to the maneuvering mode obtained in step 4 and the AIS track area attributes in step 2, adaptive statistics of the target ship life cycle are performed, specifically: A. When the maneuvering mode is uniform speed or acceleration, and the AIS track area attribute is water area, the target ship life cycle t Life The adaptive statistical method comprises the following steps: Step 5A-1: Count the maximum update time interval t max :Collect n historical dynamic update trajectory points, count the n historical update cycle intervals Δt, and record the maximum update cycle interval Δt as t max ; Step 5A-2: Calculate t Life , the specific calculation formula is: t Life =min(max(μ*t max ,t Lifemin ),t Lifemax ) Where, μ is the adaptive statistical coefficient; t Lifemin is the lower threshold of the life cycle, t Lifemax is the upper threshold of the life cycle; B. When any of the following conditions A or B occurs, the adaptive statistical coefficient μ needs to be increased; where: Condition A: When the AIS track area attribute is anchorage or berth; Condition B: When the maneuvering mode is deceleration; At this time, the target ship life cycle t Life The calculation formula is: t Life =min(max(μ'*t max ,t Lifemin ),t Lifemax ) in: μ'>μ Step 6: Pre-push: When the maneuvering mode is uniform speed, after the current AIS dynamic update, the AIS dynamic information of the target ship is pre-pushed once every collection time period ΔT; wherein the collection time period ΔT is consistent with the radar data update cycle; Step 7: Adaptive statistical prediction cycle: Predict the target ship for prediction cycle t Fore The specific calculation formula is: In the formula, is the average value within n AIS update cycles, t Poremin is the lower limit threshold of the prediction period, t Poremax is the upper threshold of the pre-push period; Step 8, end pre-pushing: For the target ship of pre-pushing, if any of the following conditions occurs during the pre-pushing process, the pre-pushing will be ended, specifically: Condition 1: The pre-pushing time reaches the pre-pushing period t calculated in step 7 Fore ; Condition 2: The geographical element area attribute of the target location predicted by the AIS track is located on land; Step 9: AIS static information update and storage.

2. The AIS track data processing method for real-time monitoring of ship status according to claim 1 is characterized in that: In step 2, the abnormal update data includes AIS jump abnormal data, AIS fallback abnormal data and AIS repeated update data.

3. The AIS track data processing method for real-time monitoring of ship status according to claim 2 is characterized in that: In step 2, the method for removing AIS jump abnormal data includes the following steps: Step 2A-1: Set the AIS jump abnormality adaptive threshold D m , the specific calculation formula is: D m =ζ*max(V0,V1)*(T1-T0) Where, ζ is the AIS position jump anomaly coefficient, which is the set value; V0 and T0 are the speed value and update time value of the AIS target in the last dynamic update data; V1 and T1 are the speed value and update time value of the AIS target in the current dynamic update data; Step 2A-2: Calculate the distance D between the last dynamic update time and the current dynamic update time of the AIS target is ; Step 2A-3, AIS jump abnormality judgment: D is With D m For comparison, when D is >D m When the AIS dynamic update data is abnormal, it is judged that the current AIS dynamic update data is abnormal and is removed.

4. The AIS track data processing method for real-time monitoring of ship status according to claim 3 is characterized in that: According to the maneuvering mode obtained in step 4, the AIS position jump anomaly coefficient ζ is automatically adjusted, specifically: A. When the maneuvering mode is uniform speed or deceleration, the influence of wind speed and water flow is taken into consideration to increase fault tolerance, and ζ takes the conventional setting value; B. When the maneuvering mode is acceleration, the AIS position jump anomaly coefficient ζ adaptively adjusts the acceleration a of the target ship. Assuming the adaptively adjusted AIS position jump anomaly coefficient is ξ', the specific calculation formula of ξ' is: ξ'>ξ,ξ'=(1+a)*ξ.

5. The AIS track data processing method for real-time monitoring of ship status according to claim 2 is characterized in that: In step 2, the method for removing abnormal data from AIS rollback includes the following steps: Step 2B-1, calculate the turning degree C m :Set N AIS dynamic update trajectory points, collect and calculate the turning degree C of the AIS target and the historical update trajectory points m ; Step 2B-2, AIS fallback abnormality judgment: turn the steering angle C m The steering threshold C hold For comparison; when C m >C hold , it is judged as AIS rollback abnormality and the AIS update data is discarded.

6. The AIS track data processing method for real-time monitoring of ship status according to claim 2 is characterized in that: In step 2, the method for eliminating AIS duplicate update data is as follows: set the speed value of the AIS target's last dynamic update data to V0, the update time to T0, and the longitude to L on0 , latitude is L at0 In this dynamic update, the speed value is V1, the update time is T1, and the longitude is L on1 , latitude is L at1 ; When (T1-T0)<T d 、V1>0 and L on1 =L on0 ,L at1 =L at0 , the AIS update data is considered to be repeated and the dynamic update time data is removed.

7. The AIS track data processing method for real-time monitoring of ship status according to claim 1 is characterized in that: Step 3: A method for determining the attributes of the AIS track area includes the following steps: Step 3-1, divide into large grids: divide the monitored sea area into large grids of size a*a; Step 3-2, export layer geographic elements: download all layer geographic elements with regional attribute information within the monitored sea area from the nautical chart; Step 3-3, primary grid refinement: each large grid that has an intersection relationship with the geographical elements of the layer exported in step 3-2 is divided into four equal parts and refined evenly to form a primary small grid refinement; Step 3-4, secondary refinement of the grid: each primary refinement grid that has an intersection relationship with the geographical elements of the layer exported in step 3-2 is divided into four equal parts and refined evenly to form secondary refinement grids; Step 3-5, m-times refined grid: repeat step 3-4, and refine the secondary refined small grid that has an intersection relationship with the geographical elements of the layer exported in step 3-2 m times to form m-times refined small grids; wherein the side lengths of the m-times refined small grids are all greater than the set size b; wherein a>b; Step 3-6, geographic element grid regional attribute storage: the grid codes and regional attributes of all large grids, as well as the regional attributes of the refined small grids are serialized and saved in files; Step 3-7, AIS track area attribute determination: When the AIS track is updated and predicted, the large grid code to which the target belongs is located according to the latitude and longitude information in the AIS dynamic information, and the area attribute information to which the target position finally belongs is queried in real time.

8. The AIS track data processing method for real-time monitoring of ship status according to claim 7 is characterized in that: a=2000 meters, b=10 meters.

9. The AIS track data processing method for real-time monitoring of ship status according to claim 1, characterized in that: In step 5, μ'=2μ.

10. The AIS track data processing method for real-time monitoring of ship status according to claim 1, characterized in that: In step 9, the AIS static information updating and storage method includes the following steps: Step 9-1, load local file: AIS historical static data is stored in the local file, including MMSI, English name, call sign, IMO, length and width information. After loading, AIS static data is saved to cache data with MMSI as the primary key; Step 9-2, AIS static information judgment: The AIS static information parsed by the AIS in step 1 is judged for null value and validity respectively; when the parsed static information is null, the corresponding static data is queried in the cache data of step 9-1 and used; when the parsed static information is not null and valid, the latest parsed static information is used, and the static information of the local cache data in step 9-1 is updated at the same time to ensure the validity and real-time of the static information of the local cache data; Step 9-3, update local files regularly: set the time interval T save , update the local cache static data updated in step 9-1 to the local file.

Images