Low refresh rate-oriented ship track association method and system for spaceborne radar, medium and device

Abstract

The application provides a low-refresh-rate-oriented ship track correlation method and system for a spaceborne radar, a medium and equipment, comprising the following steps: S1: the system analyzes the influence of low-refresh-rate observation conditions on the performance of traditional ship track correlation; S2: a wide / narrow-band alternating multi-shot joint interactive multi-mode processing mechanism is proposed, and track correlation matching is realized by extracting wide / narrow-band multi-dimensional features; S3: an evaluation model is constructed to quantitatively analyze the improvement effect of the method in the track correlation matching performance. The application proposes a solution framework based on wide / narrow-band alternating multi-shot joint interactive multi-mode processing, which fully utilizes the complementary advantages of wide / narrow-band detection in track correlation, effectively avoids the target confusion caused by sea peak splitting in wide-band detection, and the restriction of incomplete target information representation in narrow-band detection on correlation matching, thereby significantly improving the accuracy of track correlation and the overall robustness of the system.

Description

Technical Field

[0001] This invention relates to the field of aerospace systems technology, and more specifically, to a method, system, medium, and device for associating ship tracks with spaceborne radar at low refresh rates. Background Technology

[0002] Compared to airborne radar, spaceborne radar has the inherent advantage of "higher altitude and farther reach," and its deployment is not restricted by national borders, making it a key means of acquiring information on maritime targets in distant waters. However, spaceborne radar still faces several system-level technical challenges in continuous observation of surface ships: First, due to the high speed of the spaceborne platform, the time window available for target observation during a single overhead pass is extremely limited. To achieve high refresh rate ship observation, the satellite's revisit cycle must be significantly shortened, placing extremely high demands on orbit design and constellation configuration. Second, high refresh rate observation modes are usually accompanied by high energy consumption, while the energy capacity of satellite platforms is very limited, creating a significant contradiction. In addition, the heat generated by spaceborne radar during long-term continuous operation is difficult to dissipate effectively, and heat dissipation has become a key bottleneck restricting its performance and system reliability. To address these challenges, reducing the observation refresh rate is an effective technical approach. However, when conducting continuous observation of surface ships under low refresh rate conditions, how to accurately correlate and match the currently observed discrete target points with the corresponding continuous tracks of the target becomes a core problem that urgently needs to be solved. Specifically, a lower refresh rate will significantly widen the correlation gate for track association matching. This widening of the gate will cause two problems: first, the probability of non-observed targets appearing within the gate increases significantly; second, the likelihood of false targets generated by sea clutter falling within the gate also increases. These two factors combined severely reduce the accuracy and reliability of track association.

[0003] To address the target point trajectory association matching problem under low refresh rate, the paper "A Trajectory Association Method Based on Motion Information and Attribute Information" (CN202010477946.2) discloses a trajectory association method based on motion information and attribute information. This method uses multiple sensors to collaboratively detect and obtain the target's motion information and attribute information for trajectory association. For single sensor point trajectory association, no other method has been publicly available, either domestically or internationally. The paper "Method, Apparatus and Electronic Equipment for Correlation of Interrupted Tracks by Comparing Track Motion Features" (CN202510583984.9) discloses a method, apparatus and electronic equipment for correlation of interrupted tracks by comparing track motion features. This method uses a neural network to batch the acquired new track with the historical track, but does not address the point-to-point track correlation problem under low refresh rate conditions. The paper "A Divide and Conquer Track Correlation Method" (CN201710428018.5) discloses a divide and conquer track correlation method. This method uses a consensus region grid division method to solve the problem of high time consumption in multi-source track correlation algorithms, and performs high-precision registration offset screening on the quality of the registration offset of the coherence coefficient, but does not address the point-to-point track correlation problem under low refresh rate conditions. Summary of the Invention

[0004] To address the shortcomings of existing technologies, the purpose of this invention is to provide a spaceborne radar ship track association system, method, medium, and device for low refresh rates.

[0005] The method for associating ship tracks with spaceborne radar for low refresh rates provided by the present invention includes: Step S1: Analyze the impact of low refresh rate observations on the correlation performance of ship tracks based on narrowband features, and establish a quantitative relationship model between correlation performance and observation interval; Step S2: Construct a multi-mode processing mechanism for alternating wide and narrow bands, extract multi-dimensional features of wide and narrow bands, and perform track association matching; Step S3: Construct an evaluation model with association accuracy as the core indicator, and quantitatively analyze the improvement effect of this method on track association matching performance.

[0006] Preferably, step S1 includes: Step S1.1: Based on the state prediction covariance matrix Analyze the growth pattern of prediction error with observation interval T, where F is the state transition matrix and Q is the process noise covariance matrix. For discrete time step index; Step S1.2: Calculate the associated gate volume Relationship with observation interval T: , where n is the state dimension; Step S1.3: Calculate the expected value of the number of false targets within the gate. ,in Spatial density that is a false measurement; Step S1.4: Calculate the probability of correct association: ; in, For detection probability, To accurately measure the probability of falling into the gate, k is a proportionality constant.

[0007] Preferably, step S2 includes: In narrowband detection mode, the target state is predicted based on traditional filters, and the information between the predicted track state and the newly arrived measurement is calculated:

[0008] Where H is the observation matrix; The new information represents the difference between the state prediction value of the i-th track and the j-th measurement; This is the j-th measurement at time t; It is the state prediction value of the i-th track based on time t-1 for time t; A multi-hypothesis tracking method is adopted to handle the measurement-track association uncertainty caused by low refresh rate and dense target scene, and to generate a set of temporary tracks associated with the current measurement; In broadband detection mode, broadband multidimensional information of the target is extracted, including the target's radial size characteristics, central moment characteristics, and scattering center distribution characteristics.

[0009] Preferably, step S3 includes: constructing a joint correlation cost function that fuses wideband and narrowband features to quantify the degree of matching between the temporary track and the current measurement, wherein the joint correlation cost function is expressed as:

[0010] in, The cost of feature-based correlation matching for broadband signals, The cost of narrowband signal feature-based correlation matching, and These are weighting coefficients that are adaptively adjusted based on the signal-to-noise ratio and feature quality of the current detection mode. Based on the calculation results of the joint association cost function, a global optimal association decision is made, and the final confirmed track association results are updated and output.

[0011] The spaceborne radar ship track association system for low refresh rates provided by the present invention includes: The low refresh rate impact analysis module is used to analyze the growth law of prediction error with observation interval based on the state prediction covariance matrix, calculate the relationship between the associated gate volume and the observation interval, and calculate the expected value of the number of false targets in the gate. The wide-narrow alternating multi-shot joint interactive multi-mode processing module is connected to the low refresh rate impact analysis module. It is used to predict the target state and calculate the innovation based on the filter in the narrowband detection mode, generate a temporary track set using the multi-hypothesis tracking method, extract the broadband multi-dimensional information of the target in the broadband detection mode, and associate and match the extracted broadband multi-dimensional information with the temporary track set. The joint correlation cost calculation module, connected to the wide-narrow alternating multi-shot joint interactive multi-mode processing module, is used to construct a joint correlation cost function that integrates wide and narrow band features to quantify the degree of matching between the temporary track and the current measurement. The global optimal association decision module is connected to the joint association cost calculation module. It is used to make a global optimal association decision based on the calculation result of the joint association cost function, and update and output the track association result.

[0012] Preferably, the low refresh rate impact analysis module includes: Based on state prediction covariance matrix Analyze the growth pattern of prediction error with observation interval T, where F is the state transition matrix and Q is the process noise covariance matrix. For discrete time step index; Calculate the associated gate volume Relationship with observation interval T: Where n is the state dimension; calculate the expected value of the number of false targets within the gate. ,in Spatial density of spurious measurements; calculate the probability of correct association as follows: ,in, For detection probability, To accurately measure the probability of falling into the gate, k is a proportionality constant.

[0013] Preferably, the wide-narrow alternating multi-shot joint interactive multi-mode processing module includes: In narrowband detection mode, the target state is predicted based on traditional filters, and the information between the predicted track state and the newly arrived measurement is calculated:

[0014] Where H is the observation matrix; The new information represents the difference between the state prediction value of the i-th track and the j-th measurement; This is the j-th measurement at time t; It is the state prediction value of the i-th track based on time t-1 for time t; A multi-hypothesis tracking method is adopted to handle the measurement-track association uncertainty caused by low refresh rate and dense target scene, and to generate a set of temporary tracks associated with the current measurement; The broadband multidimensional information extracted in the broadband detection mode includes the target's radial size characteristics, central moment characteristics, and scattering center distribution characteristics.

[0015] Preferably, the joint association cost function is expressed as:

[0016] in, The cost of feature-based correlation matching for broadband signals, The cost of narrowband signal feature-based correlation matching, and These are weighting coefficients that are adaptively adjusted based on the signal-to-noise ratio and feature quality of the current detection mode.

[0017] According to the computer-readable storage medium storing a computer program provided by the present invention, when the computer program is executed by a processor, it implements the steps of the aforementioned method for associating ship tracks with spaceborne radar for low refresh rates.

[0018] The electronic device provided by the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the aforementioned method for associating ship tracks with spaceborne radar for low refresh rates.

[0019] Compared with the prior art, the present invention has the following beneficial effects: This invention introduces an alternating wideband and narrowband detection mechanism, effectively fusing the target's broadband fine features with narrowband kinematic features, significantly improving the accuracy of track association. This technology simultaneously solves the inherent technical challenges of incomplete target feature descriptions in traditional single narrowband detection mode and the susceptibility to sea spike interference leading to target trajectory splitting in single broadband detection mode. Thus, it achieves a significant improvement in both accuracy and robustness of track association performance in low data refresh rate scenarios. Attached Figure Description

[0020] Other features, objects, and advantages of the present invention will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 A flowchart of a method and system for associating ship tracks with spaceborne radar under low refresh rate conditions; Figure 2 This is a schematic diagram of the multi-hypothesis interaction, multi-mode, and multi-frame joint processing flow. Figure 3 This diagram illustrates the comparison of association accuracy under the same conditions, using only narrowband information for association and using interactive multi-mode multi-frame joint association with multiple hypotheses of wide and narrow information. Detailed Implementation

[0021] The present invention will now be described in detail with reference to specific embodiments. These embodiments will help those skilled in the art to further understand the present invention, but do not limit the invention in any way. It should be noted that those skilled in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0022] Example 1 like Figure 1 As shown, this invention provides a method for associating ship tracks with spaceborne radar at low refresh rates, comprising the following steps: Step S1: Analyze the impact of low refresh rate observations on the correlation performance of ship tracks based on narrowband features, and establish a quantitative relationship model between correlation performance and observation interval, specifically including: Step S1.1: Based on the state prediction covariance matrix Analyze the growth pattern of prediction error with observation interval T, where F is the state transition matrix and Q is the process noise covariance matrix. For discrete time step index; Step S1.2: Calculate the associated gate volume Relationship with observation interval T: , where n is the state dimension; Step S1.3: Calculate the expected value of the number of false targets within the gate. ,in Spatial density that is a false measurement; Step S1.4: Calculate the probability of correct association: ; in, For detection probability, To accurately measure the probability of falling into the gate, k is a proportionality constant.

[0023] Step S2: Construct a multi-mode processing mechanism for alternating wide and narrow bands, extracting multi-dimensional features from both wide and narrow bands for track association and matching, such as... Figure 2 Specifically, it includes: Step S2.1: In narrowband detection mode, predict the target state based on traditional filters and calculate the information between the predicted track state and the newly arrived measurement.

[0024] Where H is the observation matrix; The new information represents the difference between the state prediction value of the i-th track and the j-th measurement; This is the j-th measurement at time t; It is the state prediction value of the i-th track based on time t-1 for time t; Step S2.2: Employ a multi-hypothesis tracking method to address the measurement-track association uncertainty caused by low refresh rate and dense target scenes, and generate a set of temporary tracks that may be associated with the current measurement. ; Step S2.3: In broadband detection mode, extract broadband multidimensional information of the target, which includes at least the radial size features, central moment features, and scattering center distribution features of the target; and perform association matching between the extracted broadband multidimensional features and the temporary track set generated in step S2.2. Step S2.4: Construct a joint correlation cost function that integrates wideband and narrowband features to quantify the matching degree between the temporary track and the current measurement. The joint correlation cost function is expressed as follows:

[0025] in, The cost of feature-based correlation matching for broadband signals, The cost of narrowband signal feature-based correlation matching, and These are weighting coefficients that are adaptively adjusted based on the signal-to-noise ratio and feature quality of the current detection mode. Step S2.5: Based on the calculation results of the joint association cost function, perform a global optimal association decision, update and output the final confirmed track association results.

[0026] Step 3: Construct an evaluation model with association accuracy as the core indicator, and quantitatively analyze the improvement effect of this method on track association matching performance.

[0027] This embodiment provides a method for ship track association and matching at low refresh rates using spaceborne radar based on the fusion of wide and narrow band multi-features. Here, the performance evaluation of track association is conducted using target motion features and a combination of target motion features and wideband features. The specific implementation steps are as follows: Step 1: Constructing a simulation scenario: Ten ships are randomly distributed within a 100km × 100km sea area. The ships move at a speed of 15 knots, and the motion model conforms to the CV model. Radar observation parameters are shown in Table 1.

[0028] Table 1 Simulation Input Parameters

[0029] Step 2: Narrow-band motion characteristics are used for track association simulation. The association parameters are shown in Table 2. The accuracy of track association for the 10 ships is statistically analyzed as follows: Figure 3 As shown, it is 70.9%.

[0030] Table 2 Simulation Input Parameters

[0031] Step 3: Track association simulation is performed using narrowband motion features + wideband features. The association parameters are shown in Table 3. The accuracy of track association for the 10 ships is statistically analyzed as follows: Figure 3 As shown, it is 89.4%.

[0032] Table 3 Related Input Parameters

[0033] Under the same input conditions, the comparison shows that the correlation accuracy is improved by 18.5 percentage points. This indicates that by using a cooperative detection mode with alternating wide and narrow detection modes, extracting and comprehensively utilizing the multi-dimensional features of the target, and adopting an interactive multi-model and multi-shot joint processing scheme, the target trajectory correlation accuracy can be effectively improved during low refresh rate observations.

[0034] Example 2 This invention also provides a spaceborne radar ship track association system for low refresh rates, comprising: The low refresh rate impact analysis module is used to analyze the growth law of prediction error with observation interval based on the state prediction covariance matrix, calculate the relationship between the associated gate volume and the observation interval, and calculate the expected value of the number of false targets in the gate. The wide-narrow alternating multi-shot joint interactive multi-mode processing module is connected to the low refresh rate impact analysis module. It is used to predict the target state and calculate the innovation based on the filter in the narrowband detection mode, generate a temporary track set using the multi-hypothesis tracking method, extract the broadband multi-dimensional information of the target in the broadband detection mode, and associate and match the extracted broadband multi-dimensional information with the temporary track set. The joint correlation cost calculation module, connected to the wide-narrow alternating multi-shot joint interactive multi-mode processing module, is used to construct a joint correlation cost function that integrates wide and narrow band features to quantify the degree of matching between the temporary track and the current measurement. The global optimal association decision module is connected to the joint association cost calculation module. It is used to make a global optimal association decision based on the calculation result of the joint association cost function, and update and output the track association result.

[0035] Preferably, the low refresh rate impact analysis module includes: Based on state prediction covariance matrix Analyze the growth pattern of prediction error with observation interval T, where F is the state transition matrix and Q is the process noise covariance matrix. For discrete time step index; Calculate the associated gate volume Relationship with observation interval T: Where n is the state dimension; calculate the expected value of the number of false targets within the gate. ,in Spatial density of spurious measurements; calculate the probability of correct association as follows: ,in, For detection probability, To accurately measure the probability of falling into the gate, k is a proportionality constant.

[0036] Preferably, the wide-narrow alternating multi-shot joint interactive multi-mode processing module includes: In narrowband detection mode, the target state is predicted based on traditional filters, and the information between the predicted track state and the newly arrived measurement is calculated:

[0037] Where H is the observation matrix; The new information represents the difference between the state prediction value of the i-th track and the j-th measurement; This is the j-th measurement at time t; It is the state prediction value of the i-th track based on time t-1 for time t; A multi-hypothesis tracking method is adopted to handle the measurement-track association uncertainty caused by low refresh rate and dense target scene, and to generate a set of temporary tracks associated with the current measurement; The broadband multidimensional information extracted in the broadband detection mode includes the target's radial size characteristics, central moment characteristics, and scattering center distribution characteristics.

[0038] Preferably, the joint association cost function is expressed as:

[0039] in, The cost of feature-based correlation matching for broadband signals, The cost of narrowband signal feature-based correlation matching, and These are weighting coefficients that are adaptively adjusted based on the signal-to-noise ratio and feature quality of the current detection mode.

[0040] Those skilled in the art will understand that, in addition to implementing the system, apparatus, and their modules provided by this invention in purely computer-readable program code, the same program can be implemented in the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers by logically programming the method steps. Therefore, the system, apparatus, and their modules provided by this invention can be considered a hardware component, and the modules included therein for implementing various programs can also be considered structures within the hardware component; alternatively, modules for implementing various functions can be considered both software programs implementing the method and structures within the hardware component.

[0041] Specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the specific embodiments described above, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Unless otherwise specified, the embodiments and features described in this application can be arbitrarily combined with each other.

Claims

1. A method for associating ship tracks with spaceborne radar at low refresh rates, characterized in that, include: Step S1: Analyze the impact of low refresh rate observations on the correlation performance of ship tracks based on narrowband features, and establish a quantitative relationship model between correlation performance and observation interval; Step S2: Construct a multi-mode processing mechanism for alternating wide and narrow bands, extract multi-dimensional features of wide and narrow bands, and perform track association matching; Step S3: Construct an evaluation model with association accuracy as the core indicator, and quantitatively analyze the improvement effect of this method on track association matching performance.

2. The method for associating ship tracks with spaceborne radar for low refresh rates according to claim 1, characterized in that, Step S1 includes: Step S1.1: Based on the state prediction covariance matrix Analyze the growth pattern of prediction error with observation interval T, where F is the state transition matrix and Q is the process noise covariance matrix. For discrete time step index; Step S1.2: Calculate the associated gate volume Relationship with observation interval T: , where n is the state dimension; Step S1.3: Calculate the expected value of the number of false targets within the gate. ,in Spatial density that is a false measurement; Step S1.4: Calculate the probability of correct association: ; in, For detection probability, To accurately measure the probability of falling into the gate, k is a proportionality constant.

3. The method for associating ship tracks with spaceborne radar for low refresh rates according to claim 1, characterized in that, Step S2 includes: In narrowband detection mode, the target state is predicted based on traditional filters, and the information between the predicted track state and the newly arrived measurement is calculated: Where H is the observation matrix; The new information represents the difference between the state prediction value of the i-th track and the j-th measurement; This is the j-th measurement at time t; It is the state prediction value of the i-th track based on time t-1 for time t; A multi-hypothesis tracking method is adopted to handle the measurement-track association uncertainty caused by low refresh rate and dense target scene, and to generate a set of temporary tracks associated with the current measurement; In broadband detection mode, broadband multidimensional information of the target is extracted, including the target's radial size characteristics, central moment characteristics, and scattering center distribution characteristics.

4. The method for associating ship tracks with spaceborne radar for low refresh rates according to claim 1, characterized in that, Step S3 includes: constructing a joint correlation cost function that fuses wideband and narrowband features to quantify the degree of matching between temporary tracks and current measurements. The joint correlation cost function is expressed as follows: in, The cost of feature-based correlation matching for broadband signals, The cost of narrowband signal feature-based correlation matching, and These are weighting coefficients that are adaptively adjusted based on the signal-to-noise ratio and feature quality of the current detection mode. Based on the calculation results of the joint association cost function, a global optimal association decision is made, and the final confirmed track association results are updated and output.

5. A spaceborne radar ship track correlation system for low refresh rates, characterized in that, include: The low refresh rate impact analysis module is used to analyze the growth law of prediction error with observation interval based on the state prediction covariance matrix, calculate the relationship between the associated gate volume and the observation interval, and calculate the expected value of the number of false targets in the gate. The wide-narrow alternating multi-shot joint interactive multi-mode processing module is connected to the low refresh rate impact analysis module. It is used to predict the target state and calculate the innovation based on the filter in the narrowband detection mode, generate a temporary track set using the multi-hypothesis tracking method, extract the broadband multi-dimensional information of the target in the broadband detection mode, and associate and match the extracted broadband multi-dimensional information with the temporary track set. The joint correlation cost calculation module, connected to the wide-narrow alternating multi-shot joint interactive multi-mode processing module, is used to construct a joint correlation cost function that integrates wide and narrow band features to quantify the degree of matching between the temporary track and the current measurement. The global optimal association decision module is connected to the joint association cost calculation module. It is used to make a global optimal association decision based on the calculation result of the joint association cost function, and update and output the track association result.

6. The spaceborne radar ship track association system for low refresh rates according to claim 5, characterized in that, The low refresh rate impact analysis module includes: Based on state prediction covariance matrix Analyze the growth pattern of prediction error with observation interval T, where F is the state transition matrix and Q is the process noise covariance matrix. For discrete time step index; Calculate the associated gate volume Relationship with observation interval T: Where n is the state dimension; calculate the expected value of the number of false targets within the gate. ,in Spatial density of spurious measurements; calculate the probability of correct association as follows: ,in, For detection probability, To accurately measure the probability of falling into the gate, k is a proportionality constant.

7. The spaceborne radar ship track association system for low refresh rates according to claim 5, characterized in that, The alternating wide and narrow multi-shot joint interactive multi-mode processing module includes: In narrowband detection mode, the target state is predicted based on traditional filters, and the information between the predicted track state and the newly arrived measurement is calculated: Where H is the observation matrix; The new information represents the difference between the state prediction value of the i-th track and the j-th measurement; This is the j-th measurement at time t; It is the state prediction value of the i-th track based on time t-1 for time t; A multi-hypothesis tracking method is adopted to handle the measurement-track association uncertainty caused by low refresh rate and dense target scene, and to generate a set of temporary tracks associated with the current measurement; The broadband multidimensional information extracted in the broadband detection mode includes the target's radial size characteristics, central moment characteristics, and scattering center distribution characteristics.

8. The spaceborne radar ship track association system for low refresh rates according to claim 5, characterized in that, The joint association cost function is expressed as follows: in, The cost of feature-based correlation matching for broadband signals, The cost of narrowband signal feature-based correlation matching, and These are weighting coefficients that are adaptively adjusted based on the signal-to-noise ratio and feature quality of the current detection mode.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for associating ship tracks with spaceborne radar for low refresh rates as described in any one of claims 1 to 4.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of the method for associating ship tracks with spaceborne radar for low refresh rates as described in any one of claims 1 to 4.

Images