Ship tracking methods, devices, electronic equipment and storage media

Abstract

This invention discloses a ship tracking method, apparatus, electronic device, and storage medium. The specific scheme is as follows: At least one target ship within a preset sea area is identified, and the data to be processed for each target ship at the current moment is acquired; feature extraction processing is performed on the data to be processed for each target ship to obtain a first feature vector and a second feature vector; for at least one target ship, the first and second feature vectors corresponding to each target ship are input into the feature tracking model corresponding to the current moment to obtain a predicted feature vector for each target ship; predicted ship tracking information corresponding to the predicted feature vector is determined, and the navigation planning route of the target ship is determined based on the predicted ship tracking information. This invention improves the accuracy and efficiency of ship tracking information analysis, achieving the effect of real-time acquisition of predicted ship tracking information.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a ship tracking method, apparatus, electronic device, and storage medium. Background Technology

[0002] To ensure the smooth and safe navigation of the current vessel, the navigation information of other vessels in the surrounding sea area is usually analyzed in order to plan the current vessel's navigation route based on the analysis results.

[0003] Currently, the analysis of navigation information of surrounding vessels mainly relies on manual analysis of either Automatic Identification System (AIS) data or radar data to determine the vessel's current route. However, AIS data has a certain time lag, making it impossible to obtain real-time navigation information. Furthermore, radar data is affected by sea clutter, leading to missed and false detections. Therefore, analyzing only one type of data results in non-real-time and low-accuracy analysis. Moreover, manual analysis is not only inefficient but also subject to subjectivity. Summary of the Invention

[0004] This invention provides a ship tracking method, device, electronic device, and storage medium, which improves the accuracy and efficiency of ship tracking information analysis and achieves the effect of real-time acquisition and prediction of ship tracking information.

[0005] According to one aspect of the present invention, a ship tracking method is provided, the method comprising:

[0006] Identify at least one target vessel within a preset sea area and acquire the data to be processed for each target vessel at the current moment. The data to be processed includes first data collected by the target vessel's Automatic Identification System and second data collected by the vessel's radar equipment.

[0007] For each vessel to be tracked, feature extraction processing is performed on the first data and the second data to obtain the first feature vector corresponding to the first data and the second feature vector corresponding to the second data.

[0008] For at least one vessel to be tracked, the first feature vector and the second feature vector corresponding to each vessel to be tracked are input into the feature tracking model corresponding to the current time to obtain the predicted feature vector corresponding to each vessel to be tracked. The feature tracking model is determined by optimizing the feature tracking model corresponding to the previous time based on the observed feature vector of the previous time. The observed feature vector of the previous time is determined based on the historical vessel data of the previous time.

[0009] Based on the predicted feature vector, the predicted ship tracking information corresponding to the predicted feature vector is determined, so as to determine the navigation planning route of the target ship based on the predicted ship tracking information. The predicted ship tracking information includes the predicted geographical location information, predicted navigation speed information, and predicted heading angle information of the ship to be tracked at the next moment.

[0010] According to another aspect of the present invention, a ship tracking device is provided, the device comprising:

[0011] The data acquisition module is used to identify at least one target vessel within a preset sea area and acquire the data to be processed for each target vessel at the current moment. The data to be processed includes first data collected by the target vessel's Automatic Identification System and second data collected by the vessel's radar equipment.

[0012] The feature extraction module is used to perform feature extraction processing on the first data and the second data of each ship to be tracked, to obtain the first feature vector corresponding to the first data and the second feature vector corresponding to the second data.

[0013] The model prediction module is used to input the first feature vector and the second feature vector corresponding to each vessel to be tracked into the feature tracking model corresponding to the current time for at least one vessel to be tracked, so as to obtain the predicted feature vector corresponding to each vessel to be tracked. The feature tracking model is determined by optimizing the feature tracking model corresponding to the previous time based on the observed feature vector of the previous time. The observed feature vector of the previous time is determined based on the historical vessel data of the previous time.

[0014] The tracking information determination module is used to determine the predicted ship tracking information corresponding to the predicted feature vector based on the predicted feature vector, so as to determine the navigation planning route of the target ship based on the predicted ship tracking information. The predicted ship tracking information includes the predicted geographical location information, predicted navigation speed information, and predicted heading angle information of the ship to be tracked at the next moment.

[0015] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising:

[0016] At least one processor; and

[0017] A memory that is communicatively connected to at least one processor; wherein,

[0018] The memory stores a computer program that can be executed by at least one processor, such that the at least one processor is able to perform the ship tracking method of any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided that stores computer instructions for causing a processor to execute and implement the ship tracking method of any embodiment of the present invention.

[0020] According to another aspect of the present invention, a computer program product is provided, comprising a computer program, characterized in that the computer program, when executed by a processor, implements a ship tracking method as described in any embodiment of the present invention.

[0021] The technical solution of this invention identifies at least one target vessel within a preset sea area and acquires the data to be processed for each target vessel at the current moment, providing data support for subsequent prediction of vessel tracking information. Feature extraction processing is performed on the first and second data of each target vessel to obtain a first feature vector corresponding to the first data and a second feature vector corresponding to the second data. For at least one target vessel, the first and second feature vectors corresponding to each target vessel are input into the feature tracking model corresponding to the current moment to obtain a predicted feature vector for each target vessel. The feature tracking model is determined by optimizing the feature tracking model corresponding to the previous moment based on the observed feature vector at the previous moment, and the observed feature vector at the previous moment is determined based on historical vessel data from the previous moment. Based on this, accurate tracking of at least one target vessel is achieved. Based on the predicted feature vector, predicted vessel tracking information corresponding to the predicted feature vector is determined, and the navigation planning route of the target vessel is determined based on the predicted vessel tracking information. This invention solves the problems of non-real-time analysis results and low accuracy and efficiency of data processing in existing technologies, improving the accuracy and efficiency of vessel tracking information analysis and achieving the effect of real-time acquisition of predicted vessel tracking information.

[0022] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a flowchart of a ship tracking method provided in an embodiment of the present invention;

[0025] Figure 2 This is a flowchart of a feature tracking model determination method provided in an embodiment of the present invention;

[0026] Figure 3 This is a flowchart illustrating the method for determining ship tracking information provided in an embodiment of the present invention;

[0027] Figure 4 This is an example diagram of the feature tracking model update method provided in the embodiments of the present invention;

[0028] Figure 5 This is a schematic diagram of the structure of a ship tracking device provided in an embodiment of the present invention;

[0029] Figure 6 This is a schematic diagram of the structure of an electronic device that implements the ship tracking method of this invention. Detailed Implementation

[0030] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0032] Example 1

[0033] Figure 1 This is a flowchart of a ship tracking method provided in Embodiment 1 of the present invention. This embodiment is applicable to situations where ship information of one or more ships to be tracked is predicted at the next moment based on first data collected by an Automatic Identification System (AIS) and second data collected by a ship radar device. This method can be executed by a ship tracking device, which can be implemented in hardware and / or software and can be configured in electronic devices such as mobile phones, computers, or servers. Figure 1 As shown, the method includes:

[0034] S110. Identify at least one target vessel within a preset sea area and acquire the data to be processed for each target vessel at the current moment. The data to be processed includes first data collected by the target vessel's Automatic Identification System and second data collected by the vessel's radar equipment.

[0035] The target vessel can be a currently navigating vessel. During the target vessel's navigation, to ensure its safe and smooth passage, data from surrounding vessels can be monitored. Surrounding vessels can be at least one vessel to be tracked within a preset sea area. The preset sea area can be a range of waters centered on the target vessel with a preset radius. For example, the preset sea area could be a range of waters centered on the target vessel with a radius of 20 nautical miles. The vessels to be tracked are those within the preset sea area. Since the preset sea area can include one or more vessels to be tracked, there can be at least one vessel to be tracked.

[0036] Since the data to be processed for the vessel being tracked changes in real time, the current data can be obtained. This data includes first data and second data. The first data is the position, speed, and heading angle of the vessel being tracked, collected by the target vessel's Automatic Identification System (AIS). The second data is the position, speed, and heading angle of the vessel being tracked, collected by the target vessel's radar equipment. The AIS is a technical system used for automatic vessel identification and maritime communication. Vessels equipped with an AIS can automatically transmit their position, speed, and heading angle data to the target vessel's AIS. The radar equipment can be understood as the target vessel's radar equipment. Through the radar equipment, the position, speed, and heading angle of other vessels being tracked around the target vessel can be detected.

[0037] Specifically, at least one vessel to be tracked within a preset sea area at the current moment is identified as the target vessel. First data for each vessel to be tracked at the current moment is obtained based on the target vessel's Automatic Identification System (AIS). Second data for each vessel to be tracked at the current moment is obtained based on the target vessel's radar detection equipment. The first and second data are then used as data to be processed.

[0038] For example, let's take a preset sea area of ​​20 nautical miles around the target vessel as an example. At least one vessel to be tracked is identified within 20 nautical miles of the target vessel. At the current moment, the target vessel's Automatic Identification System (AIS) receives information such as the position, heading angle, and speed from the AIS of each vessel to be tracked, thus obtaining the first data for each vessel to be tracked. The target vessel's radar equipment detects and acquires the position, heading angle, and speed information of each vessel to be tracked, thus obtaining the second data for each vessel to be tracked. Based on this, data support is provided for subsequent vessel tracking information analysis.

[0039] S120. Perform feature extraction processing on the first data and the second data of each ship to be tracked to obtain the first feature vector corresponding to the first data and the second feature vector corresponding to the second data.

[0040] The first feature vector can be a feature vector extracted from the features of the first data. The second feature vector is a feature vector extracted from the features of the second data. Optionally, the first and second feature vectors can be bird's-eye view (BEV) feature vectors.

[0041] Specifically, for at least one vessel to be tracked, the feature extractor corresponding to the Automatic Identification System (AIS) performs feature extraction processing on the first data of each vessel, converting information such as position, speed, and heading angle from the first data into a first feature vector. Similarly, the feature extractor corresponding to the ship's radar equipment performs feature extraction processing on the second data of each vessel, converting information such as position, speed, and heading angle from the second data into a second feature vector. It should be noted that for each vessel's first data, there is a corresponding first feature vector. For each vessel's second data, there is a corresponding second feature vector.

[0042] For example, taking the feature extractor corresponding to the Automatic Identification System (AIS) as the AIS feature extractor and the feature extractor corresponding to the ship's radar equipment as the radar feature extractor, and the feature vector as the BEV feature vector, the following explanation is provided. For each ship to be tracked, the AIS feature extractor converts the position, speed, and heading angle data from the first data of the ship to be tracked into the corresponding BEV feature vector. The radar feature extractor converts the position, speed, and heading angle data from the second data of the ship to be tracked into the corresponding BEV feature vector.

[0043] In this embodiment of the invention, the first data includes the geographical location information, speed information, and heading angle information of the vessel to be tracked. The first feature vector can be determined as follows: for each vessel to be tracked, based on the geographical location information of the vessel to be tracked and the geographical location information of the target vessel in the first data, the relative distance information between the vessel to be tracked and the target vessel is determined; based on the relative distance information, and the speed and heading angle information of the vessel to be tracked, the feature parameters corresponding to the vessel to be tracked in a preset coordinate system are determined; based on the feature parameters, the first feature vector corresponding to the first data is determined.

[0044] The geographical location information can be understood as the latitude and longitude information of the vessel to be tracked. The heading speed information is the speed of the vessel to be tracked. The heading angle information can be the parameter of the angle between the vessel's direction of travel and true north. The relative distance information can be understood as the distance between the vessel to be tracked and the target vessel. Optionally, the relative distance information can include the lateral and longitudinal distances between the vessel to be tracked and the target vessel. The preset coordinate system can be a coordinate system set according to actual needs. Optionally, the preset coordinate system can be a BEV coordinate system with the target vessel as the origin. The feature parameters can be composed of relative distance information, speed information, and heading angle information. For example, the feature parameters can include (x, y, v, r), where x represents the lateral distance between the vessel to be tracked and the target vessel, y represents the longitudinal distance, v represents the speed information, and r represents the heading angle information.

[0045] Specifically, for each vessel to be tracked, based on the geographical location information of the current vessel and the geographical location information of the target vessel in the first data, the lateral distance and longitudinal distance information between the current vessel and the target vessel are determined. Using the lateral distance information, longitudinal distance information, and the current vessel's speed and heading angle information, the characteristic parameters corresponding to the current vessel in the preset coordinate system are determined. Based on the characteristic parameters, the feature vector corresponding to the first data is determined.

[0046] For example, taking the BEV coordinate system as a preset coordinate system, for each vessel to be tracked, based on the geographical location information of the current vessel and the geographical location information of the target vessel in the first data, the lateral distance information and longitudinal distance information between the current vessel and the target vessel are determined. The lateral distance information is used as the first feature parameter in the BEV coordinate system, and the longitudinal distance information is used as the second feature parameter in the BEV coordinate system. The speed information of the current vessel to be tracked is used as the third feature parameter in the BEV coordinate system, and the heading angle information of the current vessel to be tracked is used as the fourth feature parameter in the BEV coordinate system. Based on these four feature parameters, the first feature vector corresponding to the first data is determined.

[0047] Accordingly, the second data includes the geographical location information, speed information, and heading angle information of the vessel currently being tracked. The second feature vector can be determined as follows: for each vessel to be tracked, based on the geographical location information of the vessel and the target vessel in the second data, determine the relative distance information between the vessel and the target vessel; based on the relative distance information, as well as the speed and heading angle information of the vessel, determine the feature parameters corresponding to the vessel in the preset coordinate system; based on the feature parameters, determine the second feature vector corresponding to the second data.

[0048] In this embodiment of the invention, before inputting the first feature vector and the second feature vector corresponding to each ship to be tracked into the feature tracking model corresponding to the current time, a preset filter corresponding to the feature vector can be determined first. Specifically, this determination can be achieved by: determining the first feature category of the first feature vector and the second feature category of the second feature vector based on the environmental information corresponding to the current time; determining the first preset filter corresponding to the first feature category and the second preset filter corresponding to the second feature category in the feature tracking model, and then processing the first feature vector and the second feature vector based on the first preset filter and the second preset filter in the feature tracking model.

[0049] Optionally, the feature tracking model includes at least three types of preset filters: a Kalman filter, an Extended Kalman filter, and an Error-state Kalman filter. The feature tracking model can be a model used to predict the feature vector of the ship to be tracked at the next time step. The Kalman filter (KF) processes the input feature vector using the state equations of a linear system to achieve the optimal estimate of the feature vector or state at the next time step. The Extended Kalman Filter (EKF) processes the input feature vector using the state equations of a nonlinear system to achieve the optimal estimate of the feature vector or state at the next time step. The Error-state Kalman Filter (ESKF) processes the input feature vector using both the state equations and observation equations of the nonlinear system to achieve the optimal estimate of the feature vector or state at the next time step.

[0050] The environmental information can be environmental information collected by sensors on the target vessel within a predetermined sea area. For example, environmental information may include weather information and wave information. The feature categories can be pre-classified according to actual needs, based on the feature vectors corresponding to the data to be processed collected under different environmental conditions. Optionally, based on the feature categories, the feature vectors can be divided into: a first feature vector under simple environmental information, a first feature vector under complex environmental information, a second feature vector under simple environmental information, and a second feature vector under complex environmental information.

[0051] The feature tracking model includes at least three types of preset filters. The first preset filter can be a filter selected from at least three types that matches the first feature category. For example, if the first feature vector corresponding to the first feature category is a first feature vector under simple environmental information, then the first preset filter can be a Kalman filter; if the first feature vector corresponding to the first feature category is a first feature vector under complex environmental information, then the first preset filter can be an extended Kalman filter. The second preset filter can be a filter selected from at least three types that matches the second feature category. For example, if the second feature vector corresponding to the second feature category is a second feature vector under simple environmental information, then the second preset filter can be an extended Kalman filter; if the second feature vector corresponding to the second feature category is a second feature vector under complex environmental information, then the second preset filter can be an error Kalman filter.

[0052] Specifically, based on the environmental information at the current moment, a first feature category corresponding to the first feature vector and a second feature category corresponding to the second feature vector are determined. Based on the first feature category, a first preset filter corresponding to the first feature category and a second preset filter corresponding to the second feature category are selected from the feature tracking model. The first feature vector is processed based on the first preset filter, and the second feature vector is processed based on the second preset filter.

[0053] For example, referring to the above example, the first feature vector is the BEV feature vector obtained after processing the first data collected by the AIS system, and can therefore be simply referred to as the AIS feature vector. Correspondingly, the second feature vector is the BEV feature vector obtained after processing the second data collected by the ship's radar equipment, and can therefore be simply referred to as the radar feature vector.

[0054] Based on the environmental information at the current moment, the environment is determined to be either complex or simple. If the current environmental information is simple, the first feature category corresponding to the AIS feature vector is determined, and the first preset filter corresponding to the first feature category is determined to be a Kalman filter. Correspondingly, the second feature category corresponding to the radar feature vector is determined, and the second preset filter corresponding to the second feature category is determined to be an extended Kalman filter.

[0055] If the environmental information at the current moment is complex, then the first feature category corresponding to the AIS feature vector is determined, and the first preset filter corresponding to the first feature category is determined to be an extended Kalman filter. Correspondingly, the second feature category corresponding to the radar feature vector is determined, and the second preset filter corresponding to the second feature category is determined to be an error Kalman filter.

[0056] In this embodiment of the invention, the specific method for determining the feature category of a feature vector based on environmental information may be as follows: based on the environmental information corresponding to the current moment and a preset environmental assessment standard, determine the current sea state type corresponding to the environmental information; based on the current sea state type and a preset mapping relationship, determine the first feature category of the first feature vector and the second feature category of the second feature vector, wherein the preset mapping relationship is the mapping relationship between sea state type and feature category.

[0057] The preset environmental assessment criteria can be pre-set according to actual needs, used to assess whether environmental information is simple or complex. The current sea state type can be the type of sea area condition corresponding to the environmental information at the current moment. For example, if the preset environmental assessment criteria determine that the environmental information at the current moment is simple, then the current sea state type is simple sea state. Conversely, if the environmental information at the current moment is complex, then the current sea state type is complex sea state.

[0058] Specifically, based on preset environmental assessment standards, the current weather information, wave information, and other environmental information are analyzed to determine the current sea state type corresponding to the environmental information. According to the preset mapping relationship between sea state type and feature category and the current sea state type, the first feature category corresponding to the first feature vector is determined. Correspondingly, based on the preset mapping relationship and the current sea state type, the second feature category corresponding to the second feature vector is determined. Based on this, it is convenient to subsequently determine the preset filter corresponding to the feature category, thereby processing the feature vector based on the preset filter.

[0059] S130. For at least one vessel to be tracked, input the first feature vector and the second feature vector corresponding to each vessel to be tracked into the feature tracking model corresponding to the current time to obtain the predicted feature vector corresponding to each vessel to be tracked. The feature tracking model is determined by optimizing the feature tracking model corresponding to the previous time based on the observed feature vector of the previous time. The observed feature vector of the previous time is determined based on the historical vessel data of the previous time.

[0060] The feature tracking model can be a model used to predict the feature vector of the vessel to be tracked at the next moment. The predicted feature vector is the feature vector of the vessel to be tracked at the next moment predicted by the feature tracking model. The feature tracking model is a model that is continuously optimized and updated. The observed feature vector at the previous moment is the feature vector obtained by extracting features from the historical vessel data at the previous moment. Historical vessel data includes the first historical data collected by the Automatic Identification System (AIS) at the previous moment and / or the second historical data collected by the ship's radar equipment at the previous moment. The first historical data may include the geographical location, speed, and heading angle information of the vessel to be tracked collected by the AIS at the previous moment. The second historical data may include the geographical location, speed, and heading angle information of the vessel to be tracked collected by the ship's radar equipment at the previous moment.

[0061] Specifically, for at least one vessel to be tracked, the first and second feature vectors corresponding to each vessel are input into the feature tracking model at the current time to obtain the predicted feature vector for each vessel. Based on this, the predicted feature vector for each vessel is obtained. It should be noted that the feature tracking model at the current time is determined by optimizing the feature tracking model corresponding to the previous time based on the observed feature vector at the previous time. The observed feature vector at the previous time is determined based on the historical vessel data at the previous time.

[0062] For example, a feature tracking model can process the feature vectors of multiple ships to be tracked. If there are 10 ships to be tracked, the first and second feature vectors of these 10 ships can be input into the feature tracking model at the current moment. The kinematic matrix in the feature tracking model processes the input feature vectors to obtain the predicted feature vectors of these 10 ships at the next moment. Based on the predicted feature vectors, the geographical location, speed, and heading angle of these 10 ships at the next moment can be determined.

[0063] S140. Based on the predicted feature vector, determine the predicted ship tracking information corresponding to the predicted feature vector, so as to determine the navigation planning route of the target ship based on the predicted ship tracking information. The predicted ship tracking information includes the predicted geographical location information, predicted navigation speed information and predicted heading angle information of the ship to be tracked at the next moment.

[0064] The predicted geographical location information can be understood as the latitude and longitude information of the vessel to be tracked at the next moment. The predicted speed information can be understood as the speed information of the vessel to be tracked at the next moment. The predicted heading angle information can be understood as the parameter information of the angle between the vessel's direction of travel and true north at the next moment. The planned navigation route can be the navigation route of the target vessel determined based on the predicted vessel tracking information of at least one vessel to be tracked.

[0065] Specifically, by predicting feature vectors, the predicted geographical location, predicted speed, and predicted heading angle of each vessel to be tracked at the next moment are determined, and these information are used as the predicted vessel tracking information. Based on the predicted vessel tracking information for each vessel, the planned navigation route of the target vessel is determined.

[0066] For example, in conjunction with the above example, based on the predicted feature vectors of the 10 vessels to be tracked at the next moment, the predicted geographical location, predicted speed, and predicted heading angle of each vessel at the next moment are determined. Based on the predicted geographical location, predicted speed, and predicted heading angle of the 10 vessels to be tracked, the downstream control task of the target vessel is guided to determine the safe navigation planning route of the target vessel.

[0067] The technical solution of this embodiment identifies at least one target vessel within a preset sea area and acquires the data to be processed for each target vessel at the current moment, providing data support for subsequent prediction of vessel tracking information. Feature extraction processing is performed on the first and second data of each target vessel to obtain a first feature vector corresponding to the first data and a second feature vector corresponding to the second data. For at least one target vessel, the first and second feature vectors corresponding to each target vessel are input into the feature tracking model corresponding to the current moment to obtain a predicted feature vector for each target vessel. The feature tracking model is determined by optimizing the feature tracking model corresponding to the previous moment based on the observed feature vector at the previous moment, and the observed feature vector at the previous moment is determined based on historical vessel data from the previous moment. Based on this, accurate tracking of at least one target vessel is achieved. Based on the predicted feature vector, predicted vessel tracking information corresponding to the predicted feature vector is determined, and the navigation planning route of the target vessel is determined based on the predicted vessel tracking information. This invention solves the problems of non-real-time analysis results and low accuracy and efficiency of data processing in the prior art, improving the accuracy and efficiency of vessel tracking information analysis and achieving the effect of real-time acquisition of predicted vessel tracking information.

[0068] Example 2

[0069] Figure 2 This is a flowchart of a feature tracking model determination method provided in Embodiment 2 of the present invention. This embodiment, based on the above embodiments, requires determining the feature tracking model corresponding to the current time step before processing the feature vector using the feature tracking model corresponding to the current time. Specific implementation details can be found in the technical solution of this embodiment. Technical terms that are the same as or corresponding to those in the above embodiments will not be repeated here. Figure 2 As shown, the method includes:

[0070] S210. Obtain historical ship data of at least one ship to be tracked at the previous moment, wherein the historical ship data includes historical first data collected by the Automatic Identification System and / or historical second data collected by the ship's radar equipment.

[0071] The first historical data may include the target vessel's geographic location, speed, and heading angle information collected by the target vessel's Automatic Identification System (AIS) at the previous moment. The second historical data may include the target vessel's geographic location, speed, and heading angle information collected by the target vessel's radar equipment at the previous moment.

[0072] Specifically, at least one vessel to be tracked that was within the preset sea area of ​​the target vessel at the previous moment is identified. Historical first data of the at least one vessel to be tracked at the previous moment is obtained from the target vessel's Automatic Identification System (AIS). Historical second data of the at least one vessel to be tracked at the previous moment is obtained from the target vessel's radar equipment. The historical first data and / or historical second data are used as historical vessel data to update the model parameters in the feature tracking model.

[0073] It should be noted that when optimizing the feature tracking model, the historical ship data obtained can be the first historical data, the second historical data, or both. Different historical ship data can be used to optimize and update the corresponding preset filters in the feature tracking model.

[0074] For example, in conjunction with the above examples, see Figure 3 When optimizing and updating the feature tracking model from the previous moment, at least one vessel to be tracked within 20 nautical miles of the target vessel from the previous moment can be identified first, and historical first data and / or historical second data for each vessel to be tracked can be obtained. The historical first data corresponds to... Figure 3 The AIS data in the data. The second historical data corresponds to... Figure 3 Radar data from [the source].

[0075] S220. Perform feature extraction processing on historical ship data to determine the observation feature vector corresponding to the ship to be tracked.

[0076] Among them, the observation feature vector can be the feature vector obtained by extracting features from the historical ship data of the previous moment.

[0077] Specifically, if the historical ship data consists of historical first data collected by the Automatic Identification System (AIS), then the feature extractor corresponding to the AIS is used to extract features from the historical first data to obtain the observation feature vector corresponding to the historical first data. If the historical ship data consists of historical second data collected by the ship's radar equipment, then the feature extractor corresponding to the ship's radar equipment is used to extract features from the historical second data to obtain the observation feature vector corresponding to the historical second data. Correspondingly, if the historical ship data includes historical first data collected by the AIS and historical second data collected by the ship's radar equipment, then the corresponding feature extractors are used to obtain the observation feature vectors corresponding to the historical first data and the historical second data, respectively.

[0078] For example, see Figure 3Referring to the above examples, the AIS feature extractor corresponds to the feature extractor for the Automatic Identification System (AIS) mentioned in the above embodiments. The radar feature extractor corresponds to the feature extractor for the ship radar equipment mentioned in the above embodiments. The AIS feature extractor is used to extract features from the AIS data to obtain the BEV feature vector corresponding to the AIS data. The radar feature extractor is used to extract features from the radar data to obtain the BEV feature vector corresponding to the radar data. The BEV feature vectors corresponding to the AIS data and the BEV feature vectors corresponding to the radar data are used as the observation feature vectors of the ship to be tracked.

[0079] S230. Based on the ship information corresponding to the observed feature vector, a preset matching algorithm is used to match the observed feature vector with at least one historical prediction feature vector corresponding to the previous moment, and to determine the target historical prediction feature vector that matches the observed feature vector.

[0080] Here, ship information can be understood as the information about the ship to be tracked corresponding to the observed feature vector. The specific ship to be tracked can be determined through the ship information.

[0081] The preset matching algorithm can be a pre-defined algorithm used to match observed feature vectors and historically predicted feature vectors. Optionally, the preset matching algorithm can be the LAPJV algorithm (Linear Assignment Problem Jonker-Volgenant). The historically predicted feature vector can be a feature vector obtained by predicting the historical ship data from the previous time step using a feature tracking model. That is, the time step corresponding to the historically predicted feature vector is the same as the time step corresponding to the observed feature vector. For example, the observed feature vector can be the actual observed feature vector corresponding to the historical ship data at time A, and the historically predicted feature vector can be the predicted feature vector corresponding to the ship to be tracked at time A.

[0082] Since the feature tracking model is a multiple-input multiple-output model, the previous time step can include one or more historical predicted feature vectors corresponding to the ships to be tracked. The target historical predicted feature vector can be a historical predicted feature vector that matches the observed feature vector.

[0083] Specifically, based on the ship information of the observed feature vector, the ship to be tracked corresponding to the observed feature vector is determined. Based on the ship to be tracked corresponding to the observed feature vector, a preset matching algorithm is used to match the feature vector, that is, to determine the target historical prediction feature vector corresponding to the same ship to be tracked from at least one historical prediction feature vector.

[0084] For example, in conjunction with the above examples, see Figure 3Based on the ship information of the observed feature vector, the ship to be tracked corresponding to the observed feature vector is determined. If the feature tracking model has historical predicted feature vectors of 10 ships to be tracked at the previous moment, then the ship to be tracked that matches the ship to be tracked corresponding to the observed feature vector can be determined from these 10 ships, and the target historical predicted feature vector corresponding to that ship to be tracked can be determined.

[0085] S240. Based on the observed feature vector and the target's historical predicted feature vector, the model parameters of the feature tracking model are optimized to obtain the feature tracking model corresponding to the current moment.

[0086] Specifically, based on the observed feature vector and the target's historical predicted feature vector, the difference between the observed feature vector and the target's predicted feature vector is determined. The parameters of the corresponding preset filter in the feature tracking model are then adjusted based on this difference to obtain an optimized preset filter, thus resulting in an optimized feature tracking model.

[0087] In this embodiment of the invention, the specific way to optimize the feature tracking model may be as follows: determine the difference information between the observed feature vector and the target historical predicted feature vector; determine the historical sea state type corresponding to the previous moment, and determine the target filter in the feature tracking model based on the historical sea state type and the data acquisition dimension corresponding to the observed feature vector; optimize the state parameters of the target filter in the feature tracking model based on the difference information to obtain the feature tracking model corresponding to the current moment.

[0088] The difference information can be used to characterize the difference between the observed feature vector and the target's historical predicted feature vector. The historical sea state type can be used to characterize the sea state or environmental information corresponding to the previous moment. The historical sea state type can be simple or complex. There are two data acquisition dimensions: one is the acquisition of historical ship data through an Automatic Identification System (AIS), and the other is the acquisition of historical ship data through ship radar equipment. The data acquisition dimension can determine whether the acquisition device corresponding to the observed feature vector is an AIS or a ship radar device. The target filter can be a preset filter corresponding to the historical sea state type. The state parameters can be the parameter data used in the target filter to process the feature vector.

[0089] Specifically, the difference between the observed feature vector and the target's historical predicted feature vector is determined. Based on the historical sea state type of the previous moment and the data acquisition dimension corresponding to the observed feature vector, a target filter is determined from at least three preset filters in the feature tracking model. The state parameters of the target filter in the feature tracking model are optimized and adjusted based on the difference information to obtain the optimized target filter and the feature tracking model, thus obtaining the feature tracking model corresponding to the current moment. Based on this, the feature vector of the data to be used at the current moment can be processed according to the feature tracking model corresponding to the current moment to obtain the predicted feature vector.

[0090] For example, referring to the above example, if the historical sea state type is simple sea state and the observed feature vector is obtained from historical ship data collected by the Automatic Identification System (AIS), then the target filter is a Kalman filter. Accordingly, the Kalman filter is updated using the difference information between the observed feature vector and the target historical predicted feature vector. If the historical sea state type is simple sea state and the observed feature vector is obtained from historical ship data collected by the ship's radar equipment, then the target filter is an Extended Kalman filter (EPF). Accordingly, the EPF is updated using the difference information.

[0091] If the historical sea state type is complex and the observed feature vector is obtained from historical ship data collected by the Automatic Identification System (AIS), then the target filter is an Extended Kalman Filter (EPK). Accordingly, the EPK is updated using the difference information. If the historical sea state type is also complex and the observed feature vector is obtained from historical ship data collected by the ship's radar equipment, then the target filter is an Error Kalman Filter (EFK). Accordingly, the EFK is updated using the difference information.

[0092] For example, see Figure 3 and Figure 4Based on the above examples, the BEV feature vector corresponding to AIS data is simply referred to as the AIS feature vector, and the BEV feature vector corresponding to radar data is simply referred to as the radar feature vector. When the historical sea state type is simple, the AIS feature vector is a linear feature vector, and the state parameters in the Kalman filter are updated using the difference information between the AIS feature vector and the target historical prediction feature vector. When the historical sea state type is simple, the radar feature vector is a weakly nonlinear feature vector, and the extended Kalman filter is updated using the difference information between the radar feature vector and the target historical prediction feature vector. When the historical sea state type is complex, the AIS feature vector is a weakly nonlinear feature vector, and the extended Kalman filter is updated using the difference information between the AIS feature vector and the target historical prediction feature vector. When the historical sea state type is complex, the radar feature vector is a strongly nonlinear feature vector, and the error Kalman filter is updated using the difference information between the radar feature vector and the target historical prediction feature vector. Based on this, the preset filter in the feature tracking model is optimized and updated to obtain an optimized feature tracker. The predicted BEV feature vector for the next time step is obtained based on the feature tracker at the current time step, thereby guiding the downstream control task of the target vessel according to the predicted BEV feature vector.

[0093] The technical solution of this embodiment acquires historical ship data of at least one ship to be tracked at the previous moment, performs feature extraction processing on the historical ship data, and determines the observed feature vector corresponding to the ship to be tracked, providing data support for the subsequent optimization and updating of the feature tracking model. Based on the ship information corresponding to the observed feature vector, a preset matching algorithm is used to match the observed feature vector with at least one historical predicted feature vector corresponding to the previous moment, determining the target historical predicted feature vector that matches the observed feature vector. Through the observed feature vector and the target historical predicted feature vector, the model parameters of the feature tracking model are optimized to obtain the feature tracking model corresponding to the current moment. Through the above method, the preset filter in the feature tracking model can be dynamically adjusted in real time, enabling the optimized feature tracking model to operate efficiently under various environmental information, enhancing the reliability and adaptability of the feature tracking model. Based on this, it is convenient to obtain the predicted ship tracking information for the next moment based on the optimized feature tracking model, thereby improving the accuracy and efficiency of ship tracking information analysis and achieving the effect of obtaining predicted ship tracking information in real time.

[0094] Example 3

[0095] Figure 5 This is a structural schematic diagram of a ship tracking device provided in Embodiment 3 of the present invention. Figure 5As shown, the device includes: a data acquisition module 310, a feature extraction module 320, a model prediction module 330, and a tracking information determination module 340.

[0096] The data acquisition module 310 is used to identify at least one target vessel within a preset sea area and acquire the data to be processed for each target vessel at the current moment. The data to be processed includes first data collected by the target vessel's Automatic Identification System (AIS) and second data collected by the vessel's radar equipment. The feature extraction module 320 is used to perform feature extraction processing on the first and second data of each target vessel to obtain a first feature vector corresponding to the first data and a second feature vector corresponding to the second data. The model prediction module 330 is used to input the first and second feature vectors corresponding to each target vessel into the model prediction module for at least one target vessel. In the feature tracking model corresponding to the current moment, the predicted feature vector corresponding to each ship to be tracked is obtained. The feature tracking model is determined by optimizing the feature tracking model corresponding to the previous moment based on the observed feature vector of the previous moment. The observed feature vector of the previous moment is determined based on the historical ship data of the previous moment. The tracking information determination module 340 is used to determine the predicted ship tracking information corresponding to the predicted feature vector based on the predicted feature vector, so as to determine the navigation planning route of the target ship based on the predicted ship tracking information. The predicted ship tracking information includes the predicted geographical location information, predicted navigation speed information, and predicted heading angle information of the ship to be tracked at the next moment.

[0097] The technical solution of this embodiment identifies at least one target vessel within a preset sea area and acquires the data to be processed for each target vessel at the current moment, providing data support for subsequent prediction of vessel tracking information. Feature extraction processing is performed on the first and second data of each target vessel to obtain a first feature vector corresponding to the first data and a second feature vector corresponding to the second data. For at least one target vessel, the first and second feature vectors corresponding to each target vessel are input into the feature tracking model corresponding to the current moment to obtain a predicted feature vector for each target vessel. The feature tracking model is determined by optimizing the feature tracking model corresponding to the previous moment based on the observed feature vector at the previous moment, and the observed feature vector at the previous moment is determined based on historical vessel data from the previous moment. Based on this, accurate tracking of at least one target vessel is achieved. Based on the predicted feature vector, predicted vessel tracking information corresponding to the predicted feature vector is determined, and the navigation planning route of the target vessel is determined based on the predicted vessel tracking information. This invention solves the problems of non-real-time analysis results and low accuracy and efficiency of data processing in the prior art, improving the accuracy and efficiency of vessel tracking information analysis and achieving the effect of real-time acquisition of predicted vessel tracking information.

[0098] Based on the above embodiments, optionally, the first data includes the geographical location information, speed information, and heading angle information of the current vessel to be tracked. The feature extraction module includes: a distance information determination unit, used to determine the relative distance information between the current vessel to be tracked and the target vessel for each vessel to be tracked, based on the geographical location information of the current vessel to be tracked and the geographical location information of the target vessel in the first data; a feature parameter determination unit, used to determine the feature parameters corresponding to the current vessel to be tracked in a preset coordinate system based on the relative distance information, the speed information and heading angle information of the current vessel to be tracked; and a feature vector determination unit, used to determine the first feature vector corresponding to the first data based on the feature parameters.

[0099] Optionally, the device further includes: a filter type determination module, including a feature vector classification unit, used to determine a first feature category of a first feature vector and a second feature category of a second feature vector based on the environmental information corresponding to the current time; and a preset filter determination unit, used to determine a first preset filter corresponding to the first feature category and a second preset filter corresponding to the second feature category in the feature tracking model, so as to process the first feature vector and the second feature vector based on the first preset filter and the second preset filter in the feature tracking model.

[0100] Optionally, the feature vector classification unit is used to determine the current sea state type corresponding to the environmental information based on the environmental information at the current time and the preset environmental assessment standard; and to determine the first feature category of the first feature vector and the second feature category of the second feature vector based on the current sea state type and the preset mapping relationship, wherein the preset mapping relationship is the mapping relationship between sea state type and feature category.

[0101] Optionally, the device further includes: a model optimization module, which includes: a historical data acquisition unit, used to acquire historical ship data of at least one ship to be tracked at the previous moment, wherein the historical ship data includes historical first data collected by the Automatic Identification System and / or historical second data collected by the ship radar equipment; an observation feature vector determination unit, used to perform feature extraction processing on the historical ship data to determine the observation feature vector corresponding to the ship to be tracked; a vector matching unit, used to perform matching processing on the observation feature vector and at least one historical predicted feature vector corresponding to the previous moment based on the ship information corresponding to the observation feature vector using a preset matching algorithm to determine the target historical predicted feature vector that matches the observation feature vector; and a model optimization unit, used to optimize the model parameters of the feature tracking model based on the observation feature vector and the target historical predicted feature vector to obtain the feature tracking model corresponding to the current moment.

[0102] Optionally, a model optimization unit is used to determine the difference information between the observed feature vector and the target historical predicted feature vector; determine the historical sea state type corresponding to the previous moment, and determine the target filter in the feature tracking model based on the historical sea state type and the data acquisition dimension corresponding to the observed feature vector; and optimize the state parameters of the target filter in the feature tracking model based on the difference information to obtain the feature tracking model corresponding to the current moment.

[0103] Optionally, the feature tracking model includes at least three types of preset filters: Kalman filter, extended Kalman filter, and error Kalman filter.

[0104] The ship tracking device provided in the embodiments of the present invention can execute the ship tracking method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the method execution.

[0105] Example 4

[0106] Figure 6This is a schematic diagram of the structure of an electronic device provided in Embodiment 4 of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (such as helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0107] like Figure 6 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0108] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0109] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as ship tracking methods.

[0110] In some embodiments, the ship tracking method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the ship tracking method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the ship tracking method by any other suitable means (e.g., by means of firmware).

[0111] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0112] Computer programs for implementing the ship tracking method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The computer programs can be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0113] Example 5

[0114] Embodiment 5 of the present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to execute a ship tracking method, the method comprising:

[0115] The system identifies at least one target vessel within a preset sea area and acquires data to be processed for each target vessel at the current moment. This data includes first data collected by the target vessel's Automatic Identification System (AIS) and second data collected by the vessel's radar equipment. Feature extraction is performed on the first and second data for each target vessel to obtain a first feature vector corresponding to the first data and a second feature vector corresponding to the second data. For at least one target vessel, the first and second feature vectors are input into the feature tracking model corresponding to the current moment to obtain a predicted feature vector for each target vessel. The feature tracking model is determined by optimizing the previous moment's feature tracking model based on the observed feature vector. The observed feature vector at the previous moment is determined based on historical vessel data from the previous moment. Based on the predicted feature vector, predicted vessel tracking information is determined to inform the target vessel's navigation route. This predicted vessel tracking information includes the target vessel's predicted geographical location, predicted speed, and predicted heading angle at the next moment.

[0116] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0117] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0118] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0119] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0120] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0121] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A ship tracking method, characterized by, The method comprises the following steps: determining at least one target ship to be tracked in a preset sea area range of a target ship, and obtaining to-be-processed data of each target ship to be tracked at a current time, wherein the to-be-processed data comprises first data collected by a ship automatic identification system of the target ship and second data collected by a ship radar device; performing feature extraction processing on the first data and the second data of each target ship to be tracked respectively to obtain a first feature vector corresponding to the first data and a second feature vector corresponding to the second data; for at least one target ship to be tracked, inputting the first feature vector and the second feature vector corresponding to each target ship to be tracked into a feature tracking model corresponding to the current time to obtain a predicted feature vector corresponding to each target ship to be tracked, wherein the feature tracking model is determined by optimizing a feature tracking model corresponding to a previous time based on an observed feature vector of the previous time, and the observed feature vector of the previous time is determined based on historical ship data of the previous time; based on the predicted feature vector, determining predicted ship tracking information corresponding to the predicted feature vector to determine a navigation planning route of the target ship based on the predicted ship tracking information, wherein the predicted ship tracking information comprises predicted geographic position information, predicted navigation speed information and predicted heading angle information of the target ship to be tracked at a next time; the method further comprises: obtaining historical ship data of at least one target ship to be tracked at a previous time, wherein the historical ship data comprises historical first data collected by the ship automatic identification system and / or historical second data collected by the ship radar device; performing feature extraction processing on the historical ship data to determine an observed feature vector corresponding to the target ship to be tracked; based on ship information corresponding to the observed feature vector, performing matching processing on the observed feature vector and at least one historical predicted feature vector corresponding to the previous time by using a preset matching algorithm to determine a target historical predicted feature vector matched with the observed feature vector; based on the observed feature vector and the target historical predicted feature vector, optimizing model parameters of the feature tracking model to obtain a feature tracking model corresponding to the current time; the method further comprises: determining difference information between the observed feature vector and the target historical predicted feature vector; determining a historical sea state type corresponding to the previous time, and determining a target filter in the feature tracking model according to the historical sea state type and a data collection dimension corresponding to the observed feature vector; based on the difference information, optimizing state parameters of the target filter in the feature tracking model to obtain a feature tracking model corresponding to the current time.

2. The method of claim 1, wherein, The first data includes geographical position information, sailing speed information and heading angle information of a current ship to be tracked, feature extraction processing is performed on the first data of each ship to be tracked, and a first feature vector corresponding to the first data is determined, including: For each ship to be tracked, relative distance information between the current ship to be tracked and the target ship is determined according to the geographical position information in the first data of the current ship to be tracked and the geographical position information of the target ship; Based on the relative distance information, the sailing speed information and the heading angle information of the current ship to be tracked, a feature parameter corresponding to the current ship to be tracked in a preset coordinate system is determined; Based on the feature parameter, a first feature vector corresponding to the first data is determined.

3. The method of claim 1, wherein, Before the first feature vector and the second feature vector corresponding to each ship to be tracked are input into the feature tracking model corresponding to the current time, the method further includes: According to the environmental information corresponding to the current time, a first feature category of the first feature vector and a second feature category of the second feature vector are determined; A first preset filter corresponding to the first feature category and a second preset filter corresponding to the second feature category in the feature tracking model are determined, so that the first feature vector and the second feature vector are processed according to the first preset filter and the second preset filter in the feature tracking model.

4. The method of claim 3, wherein, According to the environmental information corresponding to the current time, a first feature category of the first feature vector and a second feature category of the second feature vector are determined, including: According to the environmental information corresponding to the current time and a preset environmental evaluation standard, a current sea state type corresponding to the environmental information is determined; According to the current sea state type and a preset mapping relationship, the first feature category of the first feature vector and the second feature category of the second feature vector are determined, wherein the preset mapping relationship is a mapping relationship between sea state types and feature categories.

5. The method of claim 1, wherein, The feature tracking model includes at least three types of preset filters, and the three types are Kalman filter, extended Kalman filter and error Kalman filter.

6. A vessel tracking apparatus characterised in that, It includes: A data acquisition module is configured to determine at least one ship to be tracked of a target ship in a preset sea area range, and acquire processed data of each ship to be tracked at a current time, wherein the processed data includes first data collected by a ship automatic identification system of the target ship and second data collected by a ship radar device; A feature extraction module is configured to perform feature extraction processing on the first data and the second data of each ship to be tracked respectively, to obtain a first feature vector corresponding to the first data and a second feature vector corresponding to the second data; a model prediction module configured to input a first feature vector and a second feature vector corresponding to each of the to-be-tracked ships into a feature tracking model corresponding to a current time, to obtain a predicted feature vector corresponding to each of the to-be-tracked ships, wherein the feature tracking model is determined by optimizing a feature tracking model corresponding to a previous time based on an observation feature vector of the previous time, and the observation feature vector of the previous time is determined based on historical ship data of the previous time; a tracking information determination module configured to determine predicted ship tracking information corresponding to the predicted feature vector based on the predicted feature vector, and to determine a navigation planning route of the target ship based on the predicted ship tracking information, wherein the predicted ship tracking information includes predicted geographic position information, predicted navigation speed information, and predicted heading angle information of the to-be-tracked ships at a next time; the device further comprises a model optimization module; the model optimization module comprises: a historical data acquisition unit configured to acquire historical ship data of at least one of the to-be-tracked ships at a previous time, wherein the historical ship data includes historical first data collected by the ship automatic identification system and / or historical second data collected by the ship radar device; an observation feature vector determination unit configured to perform feature extraction processing on the historical ship data, to determine an observation feature vector corresponding to the to-be-tracked ships; a vector matching unit configured to perform matching processing on the observation feature vector and at least one historical predicted feature vector corresponding to a previous time by using a preset matching algorithm based on ship information corresponding to the observation feature vector, to determine a target historical predicted feature vector matched with the observation feature vector; a model optimization unit configured to optimize model parameters of the feature tracking model based on the observation feature vector and the target historical predicted feature vector, to obtain a feature tracking model corresponding to a current time; the model optimization unit is specifically configured to determine difference information between the observation feature vector and the target historical predicted feature vector; determine a target filter in the feature tracking model according to a historical sea state type corresponding to the previous time and a data acquisition dimension corresponding to the observation feature vector; optimize state parameters of the target filter in the feature tracking model based on the difference information, to obtain a feature tracking model corresponding to a current time.

7. An electronic device, comprising: the electronic device comprises: at least one processor; and a memory connected with the at least one processor in communication; wherein the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to execute the ship tracking method in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, the computer readable storage medium stores computer instructions, and the computer instructions are used to enable the processor to execute the ship tracking method in any one of claims 1-5 when executed.

Images