A satellite-based real-time ship target platform dynamic association identification method

Abstract

The present invention relates to the field of target platform identification technology, and provides a satellite-based real-time ship target platform dynamic association identification method, which includes the steps of reading the configuration data of the ith task; obtaining a number of radiation source data in real time; selecting the radiation source data matching the target platform from step three according to the target platform radiation source configuration information, and generating the point track data of the ith task based on this; and associating the point track data with the historical track data to identify the target platform. The present invention performs preliminary rapid association based on the position information of the detected ship target, and realizes rapid automatic identification and processing of the platform under the situation of a wide variety of target platforms; solves the problem of poor real-time processing performance caused by the large dimension of the platform association decision matrix and the huge matrix calculation amount involved in the algorithm, and improves the recognition efficiency and effectiveness.

Description

Technical Field

[0001] The invention belongs to the technical field of target platform identification, and in particular relates to a satellite-based real-time ship target platform dynamic association identification method. Background Art

[0002] At present, my country's electronic reconnaissance satellites can form a space-based wide-area continuous reconnaissance situation during operation, and conduct long-term dynamic monitoring of key maritime ship targets. The real-time ship target platform dynamic association recognition technology based on the satellite is the core technology to achieve this function, so the research is of great significance. The research on this technology can, on the one hand, realize the motion trajectory tracking across multiple time periods for key targets, and on the other hand, provide important technical support for the construction of multiple target motion situation libraries, which is conducive to higher-precision target platform identification.

[0003] At present, there are many studies on ship target recognition based on satellite electronic information. There are two main directions for target association methods. The first is the association method based on target attribute information, and the second is the association method based on target position information. Among them, the core idea of ​​the association method based on target attributes is to compare the matching degree of target attributes as an association judgment. In addition to position and motion information, the target information also includes electromagnetic characteristic information of the target radiation source detected by the electromagnetic sensor. Therefore, the use of attribute information can solve the uncertainty problem of position information in the association of some targets. The association method based on target position information is an intuitive association method that makes judgments based on the similarity of spatial position information. Position information generally refers to the direction and motion state of the target. According to the application scenario, the characteristics of the observed data and the number of associated targets, there are usually methods such as the nearest neighbor algorithm and the multi-hypothesis tracking algorithm.

[0004] Although there have been a lot of research and achievements in many target association fields, there are still many difficulties in ship target association research based on electronic information. Mainly manifested in:

[0005] First, electromagnetic detection signal data is relatively dense, multiple observation data of the same target are obtained in a short period of time, and there are false targets. Therefore, how to extract valid targets from massive observation data containing false alarms is a great challenge;

[0006] Secondly, there is uncertainty in the detection data's judgment of the target's identity, which can lead to incorrect association of ship targets; the uncertainty of the attribute information itself can lead to inaccurate target association, etc.

[0007] In practical applications, different target platforms are equipped with different radars or communication radiation sources due to different missions and functions. For a specific target platform, its assigned radar or communication radiation source has certain rules. This rule can be used to more accurately convert the identification results of radar or communication radiation sources into the discrimination of the target platform. At the same time, the time of satellite detection of electromagnetic information of radiation sources is discontinuous and the time interval is relatively random, but for targets with not very long time intervals, their motion trajectories have certain relevance. On the basis of preliminary rapid association based on the position information of the detected ship target, the radiation source configuration information of the associated target is used for more accurate identification or manual confirmation, and dynamic association identification is continuously performed in a recursive learning manner. In this way, the dimension of the platform association decision matrix is ​​too large in the process of directly using feature information association, the matrix calculation amount involved in the algorithm is huge, resulting in poor real-time processing performance of the associated platform identification, or the platform configuration knowledge granularity is too coarse, and the accuracy of association identification is reduced. Summary of the invention

[0008] The purpose of the present invention is to provide a satellite-based real-time ship target platform dynamic association identification method to solve the problems raised in the above background technology.

[0009] A satellite-based real-time ship target platform dynamic association identification method, comprising:

[0010] Step 1: Obtain the i-th (i=1, 2, 3...n) target platform associated identification task;

[0011] Step 2: read the configuration data of the i-th task, wherein the configuration data includes the configuration information of the target platform radiation source and the historical track data of the i-th task;

[0012] Step 3: Acquire data of several radiation sources in real time;

[0013] Step 4: Filter out the radiation source data matching the target platform from step 3 according to the radiation source configuration information of the target platform, and generate the point trace data of the i-th task based on it;

[0014] Step 5: Associate the point track data with the historical track data to identify the target platform;

[0015] Step 6: Manually confirm the recognition result. If the recognition result is confirmed to be wrong, return to step 1 for re-recognition; if the recognition result is confirmed to be correct, proceed to step 7;

[0016] Step 7: Output the data of the target platform and update the historical track data;

[0017] Step 8: Repeat steps 1 to 7 to execute tasks i+1, i+2, ... n until completion.

[0018] Furthermore, the specific process of step three includes:

[0019] The data corresponding to the i-th task is stored in a two-dimensional array h. Each row of the two-dimensional array h corresponds to each radiation source detected in real time, and the column corresponds to the serial number, task number, model, location, time, and platform model of each radiation source.

[0020] Furthermore, the specific process of step 4 includes:

[0021] S1: According to the radiation source configuration information of the target platform of the i-th task, multiple radiation sources are associated with the target platform to generate a new two-dimensional array H; each row of the new two-dimensional array H corresponds to each original radiation source, and the column corresponds to the serial number of each radiation source, the serial number of the associated platform, the radiation source model, the radiation source location, time, and the associated platform model;

[0022] S2: Generate the point trace of the target platform of the i-th task at the corresponding time according to the new two-dimensional array H.

[0023] Furthermore, the specific process of step five includes:

[0024] P1: Set the ship motion speed threshold Vmax according to the historical track data of the target platform;

[0025] P2: Analyze the radiation source point trajectory of the i-th task and associate it according to the principle that the position in the corresponding time interval conforms to the normal movement speed of the ship.

[0026] Further, in P2, the specific selection process is as follows:

[0027] According to the latitude and longitude distance of the target platform with an interval time of ΔT, estimate the speed V of the target in the corresponding time interval BL , as shown below:

[0028]

[0029] Among them, B1 and B2 are the longitudes of the target platform at time T1 and T2 respectively, L1 and L2 are the latitudes of the target platform at time T1 and T2 respectively, and ΔT = T1-T2;

[0030] If V BL <Vmax, the association is successful; if V BL >Vmax, the association fails.

[0031] Furthermore, in step seven, the output target platform data includes the number of radiation sources in the new two-dimensional array H, the target platform position, and the target platform time.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] First, this method aims to solve the problem that the amount of processing calculation is too large to realize fast and automatic processing in the process of directly using feature information association when there are many types of target platforms. A real-time ship target platform dynamic association and recognition method based on satellite is proposed. The preliminary fast association is performed based on the position information of the detected ship target, so as to realize the fast and automatic recognition and processing of the platform when there are many types of target platforms.

[0034] Second, in the present invention, the real-time acquired mission data including the configuration information of the target platform radiation source and the generated dynamic association data are used as input to automatically identify the target platform and update the platform association track in real time;

[0035] Third, the on-board real-time ship target platform dynamic association identification method supports single-task input and multi-task input. For a single task, different radiation sources can be associated with different platforms, and for multiple tasks, different platforms can be associated with dynamic platform association tracks;

[0036] Fourth, the present invention realizes the association identification of the target platform of the current mission through the target platform radiation source configuration information and historical track data, solves the problems of the platform association decision matrix being too large in dimension, the huge amount of matrix calculations involved in the algorithm, resulting in poor real-time processing performance, and the platform configuration knowledge being too coarse in granularity, resulting in reduced accuracy of association identification, thereby improving the identification efficiency and effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 The present invention is a flow chart of a satellite-based real-time ship target platform dynamic association identification method. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] See also Figure 1 , a satellite-based real-time ship target platform dynamic association identification method, which includes:

[0040] Step 1: Obtain the i-th (i=1, 2, 3...n) target platform associated identification task;

[0041] Step 2: read the configuration data of the i-th task, wherein the configuration data includes the configuration information of the target platform radiation source and the historical track data of the i-th task;

[0042] Step 3: Obtain data of several radiation sources in real time; specifically including:

[0043] The data corresponding to the i-th task is stored in a two-dimensional array h. Each row of the two-dimensional array h corresponds to each radiation source detected in real time, and the column corresponds to the serial number, task number, model, location, time, and platform model of each radiation source.

[0044] Step 4: Filter out the radiation source data matching the target platform from step 3 according to the radiation source configuration information of the target platform, and generate the point trace data of the i-th task based on it; the specific process includes:

[0045] S1: According to the radiation source configuration information of the target platform of the i-th task, multiple radiation sources are associated with the target platform to generate a new two-dimensional array H; each row of the new two-dimensional array H corresponds to each original radiation source, and the column corresponds to the serial number of each radiation source, the serial number of the associated platform, the radiation source model, the radiation source location, time, and the associated platform model;

[0046] S2: Generate the point trace of the target platform of the i-th task at the corresponding time according to the new two-dimensional array H.

[0047] Step 5: Associate the point track data with the historical track data to identify the target platform; the specific process includes:

[0048] P1: Set the ship motion speed threshold Vmax according to the historical track data of the target platform;

[0049] P2: Analyze the radiation source point trajectory of the i-th task and associate it according to the principle that the position in the corresponding time interval conforms to the normal movement speed of the ship.

[0050] Among them, in step P2, the specific selection process is as follows:

[0051] According to the latitude and longitude distance of the target platform with an interval time of ΔT, estimate the speed V of the target in the corresponding time interval BL , as shown below:

[0052]

[0053] Among them, B1 and B2 are the longitudes of the target platform at time T1 and T2 respectively, L1 and L2 are the latitudes of the target platform at time T1 and T2 respectively, and ΔT = T1-T2;

[0054] If V BL <Vmax, the association is successful; if V BL>Vmax, the association fails.

[0055] Step 6: Manually confirm the recognition result. If the recognition result is confirmed to be wrong, return to step 1 for re-recognition; if the recognition result is confirmed to be correct, proceed to step 7;

[0056] Step 7: Output the target platform data and update the historical track data; the output target platform data includes but is not limited to the number of radiation sources in the new two-dimensional array H, the target platform position, and the target platform time.

[0057] Step 8: Repeat steps 1 to 7 to execute tasks i+1, i+2, ... n until completion.

[0058] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A satellite-based real-time ship target platform dynamic association identification method, characterized in that: It includes: Step 1: Obtain the i-th, i=1, 2, 3...n target platform association recognition tasks; Step 2: read the configuration data of the i-th task, wherein the configuration data includes the configuration information of the target platform radiation source and the historical track data of the i-th task; Step 3: Acquire data of several radiation sources in real time; Step 4: Filter out the radiation source data matching the target platform from step 3 according to the radiation source configuration information of the target platform, and generate the point trace data of the i-th task based on it; Step 5: Associate the point track data with the historical track data to identify the target platform; Step 6: Manually confirm the recognition result. If the recognition result is confirmed to be wrong, return to step 1 for re-recognition; if the recognition result is confirmed to be correct, proceed to step 7; Step 7: Output the data of the target platform and update the historical track data; Step 8: Repeat steps 1 to 7 to execute tasks i+1, i+2, ... n until the end; The specific process of step three includes: The data corresponding to the i-th task is stored in a two-dimensional array h, where each row of the two-dimensional array h corresponds to each radiation source detected in real time, and the column corresponds to the sequence number, task number, model, location, time, and platform model of each radiation source; The specific process of step 4 includes: S1: According to the radiation source configuration information of the target platform of the i-th task, multiple radiation sources are associated with the target platform to generate a new two-dimensional array H; each row of the new two-dimensional array H corresponds to each original radiation source, and the column corresponds to the serial number of each radiation source, the serial number of the associated platform, the radiation source model, the radiation source location, time, and the associated platform model; S2: Generate the point trace of the target platform of the i-th task at the corresponding time according to the new two-dimensional array H; The specific process of step five includes: P1: Set the ship motion speed threshold Vmax according to the historical track data of the target platform; P2: Analyze the radiation source point trace trajectory of the i-th task and associate it according to the principle that the position in the corresponding time interval conforms to the normal movement speed of the ship; In P2, the specific selection process is as follows: According to the latitude and longitude distance of the target platform with an interval time of ΔT, estimate the speed V of the target in the corresponding time interval BL , as shown below: Among them, B1 and B2 are the longitudes of the target platform at time T1 and T2 respectively, L1 and L2 are the latitudes of the target platform at time T1 and T2 respectively, and ΔT = T1-T2; If V BL <Vmax, the association is successful; if V BL >Vmax, the association fails; In step 7, the output target platform data includes the number of radiation sources in the new two-dimensional array H, the target platform position, and the target platform time.

Images