A method and system for track fusion of a navigating vessel

Abstract

This invention discloses a method and system for track fusion of ships. The method includes: first, selecting radar data and ESM data of all targets before a certain time point; performing spatial and temporal registration of the radar data and ESM data of all targets; then, calculating the similarity between data pairs based on a nearest neighbor fast discrimination method, including: distance similarity between radar data pairs, carrier frequency similarity between ESM data pairs, and azimuth similarity between radar data and ESM data; then, determining whether the data pairs correspond to the same target based on the similarity between the data pairs; if the data pairs correspond to the same target, associating the data pairs, and finally performing weighted fusion of the associating radar data and ESM data. Compared with existing technologies, this method can solve the problems of sparse data in the early stage of monitoring and the existence of detection interference and monitoring errors during the process, thereby improving the efficiency of track fusion.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method and system for fusion of the course of a ship. Background Technology

[0002] With the rapid development of science and technology, information technology has further penetrated the battlefield, and countries are paying more attention to the confrontation of information warfare. In the context of modern warfare, establishing an electronic warfare system based on information gathering and centered on information suppression has become a strategic pursuit for major military powers, both for actual combat and for the future. Due to the inherent reliability limitations of single sensors, multi-sensor fusion structures are gradually gaining attention from researchers and practical application scenarios.

[0003] Commonly used sensors include 2D radar, ESM sensors, video sensors, and sonar sensors. Researchers focus on the application of 2D radar and ESM sensors in military scenarios. ESM sensors, by intercepting and analyzing radiation source signals, obtain characteristic information of the radiation source (emission frequency, pulse width, and pulse repetition frequency, etc.), possessing strong target recognition capabilities. However, ESM sensors are essentially passive sensors and generally lack ranging capabilities. 2D radar, on the other hand, has strong target localization capabilities but weaker target recognition capabilities. Therefore, combining the target localization capabilities of 2D radar with the target recognition capabilities of ESM sensors has become an important aspect of military multi-sensor data fusion, holding significant importance in airborne early warning information processing and modern integrated electronic warfare.

[0004] However, when using existing radar and ESM sensor fusion methods, the initial data volume within the target monitoring range is very sparse, and the computational efficiency of track fusion is low. In addition, problems such as detection interference and monitoring errors exist during target monitoring, which also increases the difficulty of dynamic data fusion. Therefore, it is urgent to develop a fusion method that dynamically adjusts the target monitoring time and a track fusion method and system that improves computational efficiency. Summary of the Invention

[0005] This invention provides a method and system for fusion of ship tracks to address the problems of existing track fusion methods, such as very sparse initial data within the target monitoring range, difficulty in dynamic data correlation, detection interference and monitoring errors during the monitoring process, and low computational efficiency when correlating tracks.

[0006] The purpose of this application is to provide the following aspects:

[0007] In a first aspect, the present invention provides a method for track fusion of a navigating vessel, comprising:

[0008] The radar data and ESM data of all targets before the discrimination time point are selected. All targets include all ships that have a signal within the scanning period of the 2D radar or ESM sensor. In this invention, the scanning period refers to the scanning period of the radar or ESM sensor, which is a property of the sensor itself.

[0009] Spatial and temporal registration is performed between the radar data and ESM data of all the targets;

[0010] Based on the nearest neighbor fast discrimination method, the similarity between data pairs is calculated. The similarity between data pairs includes: distance similarity between radar data pairs, carrier frequency similarity between ESM data pairs, and azimuth similarity between radar data and ESM data.

[0011] Based on the similarity between the data pairs, it is determined whether the data pairs correspond to the same target;

[0012] If it is determined that the data pair corresponds to the same target, the data pair will be associated.

[0013] The associated radar data and ESM data are then weighted and fused.

[0014] Specifically, the spatial registration described in this invention includes: converting 2D radar data into a geocentric rectangular coordinate system; and converting 2D radar data into an ESM body coordinate system for calculation and interpolation.

[0015] The time registration described in this invention includes: since the monitored airborne ESM data cannot be used to calculate latitude and longitude coordinates, and the azimuth accuracy of the radar in this scheme is higher than that of the ESM, this scheme calibrates the measurement data to the time when the ESM provides information by interpolating the radar track; the interpolation process is based on a uniform velocity model, therefore a linear time-based interpolation method is adopted. Compared with existing technologies, this scheme can achieve real-time correlation between radar data and ESM data, effectively improving the accuracy of data correlation. Furthermore, due to the use of a parallel synchronous computing model, the overall computation time is short, improving algorithm efficiency.

[0016] Furthermore, in one implementation, the method based on rapid nearest neighbor discrimination calculates the similarity between data pairs, including:

[0017] From the radar data and ESM data of all the targets, the data pairs that need to be calculated for similarity are selected;

[0018] Calculate the similarity between data pairs at all times prior to the discrimination time point, including:

[0019] The distance similarity between radar data pairs, the carrier frequency similarity between ESM data pairs, and the azimuth similarity between radar data and ESM data are calculated for all time points prior to the discrimination time point. In this invention, "all time points prior to the discrimination time point" refers to every moment within each sensor scan cycle prior to the discrimination time point where a signal can be acquired.

[0020] Furthermore, in one implementation, determining whether a data pair corresponds to the same target based on the similarity between the data pairs includes:

[0021] The overall similarity between the data pairs is obtained by summing the similarities between the data pairs at all times before the discrimination time point;

[0022] The overall similarity between the data pairs is compared with a dynamic similarity threshold, which corresponds to the overall similarity between the data pairs. The dynamic similarity threshold includes a distance similarity threshold, a carrier frequency similarity threshold, and a location similarity threshold.

[0023] If the overall similarity between the data pairs is greater than or equal to the dynamic similarity threshold, the data pairs are determined to correspond to the same target.

[0024] In this invention, by calculating the overall similarity and then comparing the overall similarity with the dynamic similarity threshold, it is possible to determine whether data pairs belong to the same target. Compared with the prior art, this effectively reduces the problems of detection interference and monitoring errors in the monitoring process, thereby improving the accuracy of data association.

[0025] Furthermore, in one implementation, determining whether a data pair corresponds to the same target based on the similarity between the data pairs includes:

[0026] The overall distance similarity of the radar data pairs is determined based on the sum of the distance similarities between the radar data pairs at all time points; if the overall distance similarity between the radar data pairs is greater than or equal to the distance similarity threshold, the radar data pairs are determined to correspond to the same target.

[0027] The overall carrier frequency similarity of the ESM data pairs is determined based on the sum of the carrier frequency similarities between the ESM data pairs at all time points; if the overall carrier frequency similarity between the ESM data pairs is greater than or equal to the carrier frequency similarity threshold, the ESM data pairs are determined to correspond to the same target.

[0028] The overall azimuth similarity between the radar data and ESM data is determined based on the sum of the azimuth similarity between the radar data and ESM data at all time points. If the overall azimuth similarity between the radar and ESM sensors is greater than or equal to the azimuth similarity threshold, it is determined that the radar data and ESM data correspond to the same target.

[0029] In this invention, by using the overall similarity to determine whether they are the same target, the problems of detection interference and monitoring errors in the monitoring process can be effectively reduced, thereby improving the accuracy of data association.

[0030] Furthermore, in one implementation, determining whether a data pair corresponds to the same target based on the similarity between the data pairs includes:

[0031] If the discrimination time point belongs to the initial monitoring stage, the dynamic similarity threshold is set as the first threshold. If the similarity is greater than or equal to the first threshold and the similarity is less than or equal to the sensor error, the data pair is determined to correspond to the same target. Specifically, in this invention, the first threshold refers to an empirical value, which can be adjusted according to the experience of those skilled in the art. The initial monitoring stage is set before the radar or ESM sensor with the most monitoring data receives 5 signal data. The error is a given value.

[0032] If the discrimination time point does not belong to the initial monitoring stage, the dynamic similarity threshold is set as the second threshold. If the similarity is greater than or equal to the second threshold, and among the similarities corresponding to time points before the discrimination time point, the similarities corresponding to time points with a preset proportion are all greater than or equal to the second threshold, then the data pair is determined to correspond to the same target.

[0033] Furthermore, in one implementation, the initial monitoring phase is determined based on the number of acquireable targets, and the number of acquireable targets increases as the targets enter the monitoring range:

[0034] If the number of targets that can be acquired increases to less than a preset threshold as the time it takes for the target to enter the monitoring range of the radar or ESM sensor increases, then the determination time point belongs to the initial monitoring stage.

[0035] If the number of targets that can be acquired increases to a level greater than or equal to a preset threshold as the time it takes for the target to enter the monitoring range of the radar or ESM sensor increases, then the determination time point does not belong to the initial monitoring phase.

[0036] Specifically, in this invention, the "initial monitoring period" for each sensor is determined independently, and can be set according to the number of data received since the first data reception. Through the dynamic similarity threshold and the setting of the initial monitoring period, this invention enables different determinations of the similarity of data pairs based on the discrimination time point. This effectively solves the problem of very sparse initial data within the target monitoring range and the difficulty in dynamically associating data in existing track fusion methods. Therefore, compared to existing technologies, this reduces monitoring errors and improves monitoring accuracy.

[0037] Furthermore, in one implementation, the method further includes:

[0038] Before selecting radar data and ESM data for all targets at the discrimination time point, prior knowledge is acquired and used, including:

[0039] If there is no temporal overlap between the radar data and the ESM data, then the radar data and the ESM data are excluded from the candidates for similarity calculation.

[0040] If the similarity between the radar data and the ESM data has been obtained, then the similarity is cached.

[0041] If the association between the radar data and ESM data has been determined, then the associated radar data and ESM data are cached.

[0042] The radar data for each batch is matched one-to-one with each target.

[0043] In this invention, since the algorithm adopts a parallel synchronous computing model and utilizes prior knowledge to significantly reduce the number of calculations in the correlation calculation process, the efficiency of track fusion is greatly improved compared with the existing technology.

[0044] Furthermore, in one implementation, the weighted fusion of the associated radar data and ESM data includes:

[0045] The associated radar data and ESM data are weighted and fused, and the time point of the ESM data is interpolated according to the time point of radar data acquisition.

[0046] Singularities in the fused data are removed, and the fused data is then smoothed.

[0047] Secondly, the present invention provides a trajectory fusion system for a navigating vessel, comprising:

[0048] The data selection module is used to select radar data and ESM data of all targets before the judgment time point. All targets include all ships that have signals within the scanning period of the 2D radar or ESM sensor.

[0049] The registration module is used to perform spatial and temporal registration of the radar data and ESM data of all the targets.

[0050] The similarity calculation module is used to calculate the similarity between data pairs based on the nearest neighbor fast discrimination method. The similarity between the data pairs includes: the distance similarity between radar data pairs, the carrier frequency similarity between ESM data pairs, and the azimuth similarity between radar data and ESM data.

[0051] The determination module is used to determine whether the data pairs correspond to the same target based on the similarity between the data pairs;

[0052] The association module is used to associate the data pairs if it is determined that the data pairs correspond to the same target;

[0053] The fusion module is used to perform weighted fusion of the associated radar data and ESM data.

[0054] Thirdly, the present invention provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the steps of the aforementioned method for track fusion of a navigating vessel.

[0055] Furthermore, the present invention also provides a track fusion device for a navigating vessel, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor to perform a track fusion method for the navigating vessel.

[0056] As can be seen from the above technical solutions, the embodiments of the present invention provide a method and system for track fusion of a sailing vessel. The method includes: first, selecting radar data and ESM data of all targets before the discrimination time point; performing spatial and temporal registration of the radar data and ESM data of all targets; then, calculating the similarity between data pairs based on a nearest neighbor fast discrimination method, including: distance similarity between radar data pairs, carrier frequency similarity between ESM data pairs, and azimuth similarity between radar data and ESM data; then, determining whether the data pairs correspond to the same target based on the similarity between the data pairs; if the data pairs correspond to the same target, associating the data pairs, and finally performing weighted fusion of the associated radar data and ESM data.

[0057] In existing technologies, the initial data volume within the target monitoring range is very sparse, making dynamic data association difficult. Furthermore, the monitoring process suffers from detection interference, monitoring errors, and low computational efficiency when associating flight tracks. The aforementioned method, however, determines whether radar data and ESM data correspond to the same target by assessing the distance similarity between radar data pairs, the carrier frequency similarity between ESM data pairs, and the azimuth similarity between radar data and ESM data. Data corresponding to the same target are then associated and fused. Compared to existing technologies, this method achieves dynamic data association while maintaining a certain level of association accuracy, and simultaneously improves computational efficiency. Attached Figure Description

[0058] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without creative effort.

[0059] Figure 1 This is a schematic diagram of the workflow of a navigation vessel track fusion method provided in this application;

[0060] Figure 2 This is a schematic diagram of the similarity calculation workflow in a trajectory fusion method for a navigating vessel provided in this application;

[0061] Figure 3 This is a schematic diagram of the first target determination process in a trajectory fusion method for a navigating vessel provided in this application;

[0062] Figure 4 This is a schematic diagram of the second target determination process in a trajectory fusion method for a navigating vessel provided in this application;

[0063] Figure 5 This is a schematic diagram of the third target determination process in a trajectory fusion method for a navigating vessel provided in this application;

[0064] Figure 6 This is a schematic diagram of the initial monitoring setup workflow in a tracking fusion method for a navigating vessel provided in this application;

[0065] Figure 7 This is a schematic diagram of the prior knowledge setting workflow in a trajectory fusion method for a navigating vessel provided in this application;

[0066] Figure 8 This is a schematic diagram of the weighted fusion workflow in a trajectory fusion method for a navigating vessel provided in this application;

[0067] Figure 9This is a structural block diagram of a navigation track fusion system for a ship provided in this application. Detailed Implementation

[0068] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of methods consistent with some aspects of the invention as detailed in the appended claims.

[0069] The following detailed embodiments illustrate a method and system for fusion of navigation vessels provided in this application.

[0070] This invention discloses a method and system for track fusion of ships at sea. This solution is applied to real-time track fusion of maritime targets based on 2D radar data and ESM sensor data. In this embodiment, the 2D radar and ESM sensor are located at the same observation station, and the detection ranges of each radar partially overlap but do not overlap. Therefore, if a target appears within a certain detection range of each sensor, it can be detected by both the radar and the ESM sensor. That is, different radars and different ESM sensors can acquire detection data for multiple targets. At this point, it is necessary to determine whether the data detected by these radars and ESM sensors belong to the same target. If it is determined that the data belong to the same target, it is necessary to correlate the radar data and ESM data.

[0071] Those skilled in the art will recognize that a 2D radar acquires at least the following data: batch number, time, sensor number, target azimuth, and range; while an ESM sensor acquires at least the following data: batch number, time, sensor number, target azimuth, carrier frequency, repetition rate, and pulse width. Furthermore, each radar and ESM sensor inevitably introduces errors when receiving data; these errors are fixed parameters of the radar and ESM sensor.

[0072] The data detected by 2D radar and ESM sensors are relatively continuous trajectories. If signal problems or environmental interference occur during the detection process, the trajectory will have breaks. Furthermore, since both radar and ESM sensors have certain errors in data reception, data correlation becomes more difficult. In addition, the scanning cycles of various radars and ESM sensors are difficult to standardize; for example, the scanning cycles may include 5s / time, 10s / time, 20s / time, etc.

[0073] For example, see Figure 1 This diagram illustrates the workflow of a trajectory fusion method for a sailing vessel provided by an embodiment of the present invention. Combined with... Figure 1As can be seen, the trajectory fusion method for a navigating vessel described in this embodiment includes:

[0074] S101, Select radar data and ESM data of all targets before the discrimination time point, wherein all targets include all ships that have signals within the scanning period of 2D radar or ESM sensor;

[0075] S102, Spatial and temporal registration of the radar data and ESM data of all the targets;

[0076] S103, Based on the nearest neighbor fast discrimination method, calculate the similarity between data pairs. The similarity between data pairs includes: distance similarity between radar data pairs, carrier frequency similarity between ESM data pairs, and azimuth similarity between radar data and ESM data.

[0077] S104, Based on the similarity between the data pairs, determine whether the data pairs correspond to the same target;

[0078] S105, if it is determined that the data pair corresponds to the same target, the data pair is associated;

[0079] S106, the associated radar data and ESM data are weighted and fused.

[0080] In maritime target monitoring applications, the operational states of maritime targets are relatively stable. For example, the changes in the short-term motion of a ship of 1,000 tons or 10,000 tons during acceleration, deceleration, and turning are negligible. Therefore, the method described in this embodiment uses a uniform linear model as the state estimation model. Furthermore, the method described in this embodiment employs a filtering-based algorithm to smooth a series of fluctuating data.

[0081] Specifically, the similarity in the prior art includes: similarity between similar sensors and similarity between dissimilar sensors. In this embodiment, the similarity between similar sensors includes the similarity between radars and the similarity between ESM sensors, and the similarity between dissimilar sensors includes the similarity between radars and ESM sensors. The similarity between radars is defined using distance similarity. At the same time, the greater the distance between two targets, the smaller the distance similarity; the closer the two targets are, the greater the distance similarity. Specifically, the distance similarity threshold between radars is set based on empirical values, and in the actual code, it is set to 1000 meters. That is, if the difference in target distance between two radar data monitored at the same time point is less than 1000 meters, the similarity between the two radar data is considered high, indicating that the two radar data belong to the same target and can be correlated. Conversely, if the difference in target distance between two radar data monitored at the same time point is greater than or equal to 1000 meters, the similarity between the two radar data is considered low, indicating that the two radar data belong to different targets. Furthermore, in this invention, the distance between the targets is calculated based on the azimuth and distance of the targets acquired by the radar. Specifically, the coordinates of the targets can be obtained based on the azimuth and distance of the targets, and then the distance between the two targets is determined based on the coordinates of the two targets, and the distance similarity is judged.

[0082] The similarity between ESM and ESM sensors is defined using carrier frequency similarity. At the same time, the greater the difference in carrier frequencies acquired by the two ESM sensors, the smaller the carrier frequency similarity; conversely, the smaller the difference, the greater the similarity. In combat scenarios, the carrier frequency monitored by our ESM sensors is the frequency of the enemy target signal. If, at the same time, the carrier frequencies of the same enemy target detected by two ESM sensors are not significantly different, this scheme determines whether the target data acquired by the two ESM sensors at the same time belong to the same target based on carrier frequency similarity. Specifically, the carrier frequency similarity threshold between ESM sensors is determined empirically. Furthermore, it should be noted that in practical applications, each observation station typically has only one ESM sensor. When performing track association for a single observation station, the issue of data association between ESM sensors does not need to be considered. However, when implementing track association for multiple observation stations, the similarity between ESM sensors is determined based on carrier frequency rather than azimuth or other data.

[0083] The similarity between the radar and the ESM sensor is defined using azimuth similarity. A larger difference in azimuth angle between the radar and the ESM sensor indicates a smaller similarity, while a smaller difference indicates a larger similarity. Specifically, the carrier frequency similarity threshold between the radar and the ESM sensor is determined based on empirical values.

[0084] In addition to the nearest neighbor-based fast association algorithm in this scheme, radar track association can also be achieved using Mahalanobis distance, or similarity measures such as likelihood function, likelihood ratio, maximum a posteriori probability, target space topology, and target attributes, or association algorithms based on statistical probability or fuzzy analysis.

[0085] Specifically, in this invention, the similarity calculation formula is as follows:

[0086]

[0087] in, This represents the similarity between data pairs at time t;

[0088] This represents the difference between data pairs at time t, when comparing 2D radar data. This is the range difference when comparing 2D radar data and ESM data. This is the angle difference;

[0089] Based on empirical values, the final distance between 2D radar data and the angle between 2D radar data and ESM data adopted in this scheme is 2000 meters.

[0090] The formula for calculating the distance difference is:

[0091]

[0092] Where a = latitude1 - latitude2, a represents the difference in latitude between two points, latitude1 represents the latitude of the first point, and latitude2 represents the latitude of the second point; b = longitude1 - longitude2, represents the difference in longitude between two points, longitude1 represents the longitude of the first point, and longitude2 represents the longitude of the second point; 6378137 represents the Earth's radius.

[0093] The formula for calculating the angle difference is the absolute value of the difference between the two angles.

[0094] The present invention discloses a method for track fusion of ships. First, radar data and ESM data of all targets prior to the identification time point are selected. Spatial and temporal registration is performed on the radar data and ESM data of all targets. Then, based on a nearest neighbor fast discrimination method, the similarity between data pairs is calculated, including: distance similarity between radar data pairs, carrier frequency similarity between ESM data pairs, and azimuth similarity between radar data and ESM data. Next, the similarity between data pairs is used to determine whether the data pairs correspond to the same target. If the data pairs correspond to the same target, they are associated. Finally, the associated radar data and ESM data are weighted and fused. Therefore, compared with existing technologies, this method can solve the problems of sparse data in the early stages of monitoring and the existence of detection interference and monitoring errors during the process, thereby improving the efficiency of track fusion.

[0095] See Figure 2 In the trajectory fusion method for a navigating vessel described in this embodiment, the method based on nearest neighbor fast discrimination calculates the similarity between data pairs, including:

[0096] S201, from the radar data and ESM data of all targets, filter out the data pairs that need to be calculated for similarity;

[0097] S202, Calculate the similarity between data pairs at all times prior to the discrimination time point, including:

[0098] Calculate the distance similarity between radar data pairs, the carrier frequency similarity between ESM data pairs, and the azimuth similarity between radar data and ESM data for all time points prior to the discrimination time point.

[0099] In this embodiment, specifically, the screening method includes: data from different batches of the same sensor are not calculated, and the data times of candidate calculation pairs must overlap for a period of time. "All moments before the discrimination time point" refers to every moment within each sensor scanning cycle before the discrimination time point where a signal can be acquired.

[0100] See Figure 3 In the trajectory fusion method for a navigating vessel described in this embodiment, the step of determining whether a data pair corresponds to the same target based on the similarity between data pairs includes:

[0101] S301, the overall similarity between the data pairs is obtained by summing the similarities between the data pairs at all times before the discrimination time point;

[0102] S302, compare the overall similarity between the data pairs with the dynamic similarity threshold, wherein the dynamic similarity threshold corresponds to the overall similarity between the data pairs, and the dynamic similarity threshold includes a distance similarity threshold, a carrier frequency similarity threshold, and a location similarity threshold;

[0103] S303, if the overall similarity between the data pairs is greater than or equal to the dynamic similarity threshold, the data pairs are determined to correspond to the same target.

[0104] In this embodiment, by calculating the overall similarity and then comparing the overall similarity with the dynamic similarity threshold, it is possible to determine whether the data pairs belong to the same target. Compared with the prior art, this effectively reduces the problems of detection interference and monitoring errors in the monitoring process, thereby improving the accuracy of data association.

[0105] See Figure 4 In the trajectory fusion method for a navigating vessel described in this embodiment, the step of determining whether a data pair corresponds to the same target based on the similarity between data pairs includes:

[0106] S401, determine the overall distance similarity of the radar data pairs based on the sum of the distance similarities between the radar data pairs at all time points; if the overall distance similarity between the radar data pairs is greater than or equal to the distance similarity threshold, determine that the radar data pairs correspond to the same target;

[0107] S402, determine the overall carrier frequency similarity of the ESM data pairs based on the sum of the carrier frequency similarities between the ESM data pairs at all time points; if the overall carrier frequency similarity between the ESM data pairs is greater than or equal to the carrier frequency similarity threshold, determine that the ESM data pairs correspond to the same target.

[0108] S403, determine the overall azimuth similarity between the radar data and ESM data based on the sum of the azimuth similarity between the radar data and ESM data at all time points; if the overall azimuth similarity between the radar and ESM sensors is greater than or equal to the azimuth similarity threshold, determine that the radar data and ESM data correspond to the same target.

[0109] In this embodiment, by using the overall similarity to determine whether they are the same target, the problems of detection interference and monitoring errors in the monitoring process can be effectively reduced, thereby improving the accuracy of data association.

[0110] See Figure 5 In the trajectory fusion method for a navigating vessel described in this embodiment, the step of determining whether a data pair corresponds to the same target based on the similarity between data pairs includes:

[0111] S501, adjusting the dynamic similarity threshold according to the monitoring time period to which the discrimination time point belongs, including:

[0112] S502, if the discrimination time point belongs to the initial monitoring stage, the dynamic similarity threshold is set as the first threshold. If the similarity is greater than or equal to the first threshold and the similarity is less than or equal to the sensor error, the data pair is determined to correspond to the same target. Specifically, in this embodiment, the first threshold refers to an empirical value, which can be adjusted according to the experience of those skilled in the art. The initial monitoring stage is set before the radar or ESM sensor with the most monitoring data receives 5 signal data. The error is a given value.

[0113] S503, if the discrimination time point does not belong to the initial monitoring stage, the dynamic similarity threshold is set to the second threshold. If the similarity is greater than or equal to the second threshold, and among the similarities corresponding to time points before the discrimination time point, the similarities corresponding to time points with a preset proportion are all greater than or equal to the second threshold, then the data pair is determined to correspond to the same target.

[0114] Specifically, the second threshold refers to an empirical value, which can be adjusted according to the experience of those skilled in the art. In this embodiment, the second threshold can be set to 0.5.

[0115] See Figure 6 The method for track fusion of a navigating vessel described in this embodiment includes:

[0116] S601, the initial monitoring phase is determined based on the number of acquireable targets, and the number of acquireable targets increases as the target enters the monitoring range:

[0117] S602, if the number of targets that can be acquired increases to less than a preset number threshold as the time it takes for the target to enter the monitoring range of the radar or ESM sensor increases, then the discrimination time point belongs to the initial monitoring stage.

[0118] S603, if the number of targets that can be acquired increases to a level greater than or equal to a preset threshold as the time it takes for the target to enter the monitoring range of the radar or ESM sensor increases, then the determination time point does not belong to the initial monitoring stage.

[0119] Specifically, in this embodiment, the "initial monitoring period" for each sensor is determined independently, and can be set according to the number of data received since the first data reception. In this invention, by setting the dynamic similarity threshold and the initial monitoring period, different judgments can be made on the similarity of data pairs based on the judgment time point. This effectively solves the problem that existing track fusion methods have very sparse initial data within the target monitoring range, making dynamic data correlation difficult. Therefore, compared to existing technologies, this reduces monitoring errors and improves monitoring accuracy.

[0120] See Figure 7 In this embodiment, the method for fusion of the course of a navigating vessel further includes:

[0121] S701, before selecting radar data and ESM data of all targets at the discrimination time point, prior knowledge is acquired and used, the prior knowledge including:

[0122] S702, if there is no temporal overlap between the radar data and the ESM data, then the radar data and the ESM data are excluded from the candidates for which similarity calculation is required.

[0123] S703, if the similarity between the radar data and the ESM data has been obtained, then the similarity is cached;

[0124] S704, if the association between the radar data and the ESM data has been determined, then the associated radar data and ESM data are cached.

[0125] S705 maps the radar data of each batch number to each target.

[0126] In this embodiment, since the algorithm adopts a parallel synchronous computing model and utilizes prior knowledge to significantly reduce the number of calculations in the correlation calculation process, the track fusion efficiency is greatly improved compared with the existing technology.

[0127] See Figure 8 In the trajectory fusion method for a navigating vessel described in this embodiment, the weighted fusion of associated radar data and ESM data includes:

[0128] S801, the associated radar data and ESM data are weighted and fused, and the time point of the ESM data is interpolated according to the time point of radar data acquisition.

[0129] S802, remove the singular points in the fused data and smooth the fused data.

[0130] Specifically, in this embodiment, track fusion is a weighted fusion and data smoothing based on multi-sensor data. This track fusion process is based on two assumptions: the sensors are independent of each other; and the sensor errors follow a Gaussian distribution. In this embodiment, by weighted fusion of ESM track data and radar track data, and then removing singularities from the fused data, data smoothing is finally achieved on the fused data. The methods for smoothing the fused data include, but are not limited to, moving average smoothing, Savitzky-Golay filtering smoothing, and Kalman filtering smoothing.

[0131] In summary, in the trajectory fusion method for a navigating vessel provided in this embodiment, firstly, based on the latitude and longitude obtained after spatial registration of radar data trajectory points, a preliminary nearest-neighbor similarity judgment is used to obtain a list of candidate data groups to be associated and fused. Since the calculation is based on data received by each sensor over the entire time period, this method first requires that data between different sensor pairs within the common monitoring range must be synchronized for the vast majority of the time; otherwise, they do not need to be included in the candidate calculation list. The specific calculation between radar data is divided into two steps: the threshold is appropriately increased for the first round of calculation to avoid mutual interference when there are many targets; after the first round of calculation, association, and fusion are completed, for data with a small number of associations, a second association calculation is performed with the existing candidates. Specifically, the small number of associations can be set to no more than 3 sensors having data. At this time, the threshold is lowered to avoid the existence of isolated data. The aforementioned steps are a technique based on the precondition that "data with the same batch information from the same radar is generated by the same target." For those data that have been partially merged within the multi-sensor monitoring range but not fully merged, a second calculation and merging is performed, using the partially merged data in this merging calculation. Finally, the ESM data is correlated and fused to merge the tracks of ESM data and radar data: the ESM data and radar data also adopt the aforementioned nearest neighbor fast discrimination method; and for different ESM data pairs, the intersection point of multiple ESM sensor data is calculated to obtain the trajectory calculated only based on ESM data. This data can be used to connect broken candidate sets of radar data based on correlated data, thereby achieving further correlation.

[0132] Based on the aforementioned optimizations to the trajectory association and fusion calculation process, the algorithm is fully parallelized, making full use of multi-core computing resources. In subsequent practical applications, further distributed computing using a distributed cluster can be implemented to further improve efficiency. The algorithm primarily consumes CPU and memory resources. Local efficiency verification was conducted on a machine with an Intel i7-8700K 3.7GHz CPU and 16GB of memory, with the total execution time for trajectory association and fusion being approximately 2 seconds.

[0133] The various method embodiments described herein can be independent solutions or combinations based on internal logic, and all such solutions fall within the protection scope of this application.

[0134] It is understood that the methods and operations implemented by the network device in the above-described method embodiments can also be implemented by components (such as chips or circuits) that can be used in the network device.

[0135] The above embodiments describe the trajectory fusion method for ships provided in this application. It is understood that, in order to achieve the above functions, the network device includes corresponding hardware structures and / or software modules for executing each function. Those skilled in the art should readily recognize that, based on the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0136] This application embodiment can divide the network device into functional modules according to the above method example. For example, each function can be divided into a separate functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. It should be noted that the module division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0137] The above, combined with Figures 1 to 8 The methods provided in the embodiments of this application are described in detail below. Figure 9 The system provided in the embodiments of this application is described in detail. It should be understood that the description of the system embodiments corresponds to the description of the method embodiments. Therefore, for content not described in detail, please refer to the method embodiments above. For the sake of brevity, it will not be repeated here.

[0138] This embodiment also provides a trajectory fusion system for a navigating vessel, the system 900 comprising:

[0139] The data selection module 901 is used to select radar data and ESM data of all targets before the judgment time point, wherein all targets include all ships that have signals within the scanning period of 2D radar or ESM sensor;

[0140] The registration module 902 is used to perform spatial and temporal registration of the radar data and ESM data of all the targets;

[0141] The similarity calculation module 903 is used to calculate the similarity between data pairs based on the nearest neighbor fast discrimination method. The similarity between the data pairs includes: the distance similarity between radar data pairs, the carrier frequency similarity between ESM data pairs, and the azimuth similarity between radar data and ESM data.

[0142] The determination module 904 is used to determine whether the data pairs correspond to the same target based on the similarity between the data pairs;

[0143] The association module 905 is used to associate the data pairs if it is determined that the data pairs correspond to the same target;

[0144] The fusion module 906 is used to perform weighted fusion of the associated radar data and ESM data.

[0145] This embodiment also provides a computer-readable storage medium storing computer instructions that, when executed by a processor, implement the track fusion step of the aforementioned navigation vessel.

[0146] Those skilled in the art will recognize that the various illustrative logical blocks and steps described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0147] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0148] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0149] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0150] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0151] The navigation vessel track fusion system provided in the above-described embodiments of this application is used to execute the method described above. Therefore, the beneficial effects it can achieve can be referred to the beneficial effects corresponding to the method described above, and will not be repeated here.

[0152] It should be understood that in the various embodiments of this application, the execution order of each step should be determined by its function and internal logic, and the size of each step number does not mean the order of execution, and does not constitute a limitation on the implementation process of the embodiments.

[0153] The various parts of this specification are described in a progressive manner. Similar or identical parts between the various embodiments can be referred to interchangeably. Each embodiment focuses on the differences from other embodiments. In particular, the embodiments for the trajectory fusion system of a sailing vessel are basically similar to the method embodiments, so the description is relatively simple; relevant details can be found in the descriptions within the method embodiments.

[0154] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0155] The embodiments described above do not constitute a limitation on the scope of protection of this application.

Claims

1. A method for fusion of the tracks of a navigating vessel, characterized in that, include: Select radar data and ESM data of all targets prior to the identification time point, wherein all targets include all ships that have a signal within the scanning period of 2D radar or ESM sensor; Spatial and temporal registration is performed between the radar data and ESM data of all the targets; Based on the nearest neighbor fast discrimination method, the similarity between data pairs is calculated, including: From the radar data and ESM data of all the targets, the data pairs that need to be calculated for similarity are selected; Calculate the similarity between data pairs at all times before the discrimination time point, that is, calculate the distance similarity between radar data pairs, the carrier frequency similarity between ESM data pairs, and the azimuth similarity between radar data and ESM data at all times before the discrimination time point. Determining whether a data pair corresponds to the same target based on the similarity between the data pairs includes: The overall similarity between the data pairs is obtained by summing the similarities between the data pairs at all times before the discrimination time point; The overall similarity between the data pairs is compared with a dynamic similarity threshold, which corresponds to the overall similarity between the data pairs. The dynamic similarity threshold includes a distance similarity threshold, a carrier frequency similarity threshold, and a location similarity threshold. If the overall similarity between the data pairs is greater than or equal to the dynamic similarity threshold, the data pairs are determined to correspond to the same target. If it is determined that the data pair corresponds to the same target, the data pair will be associated. The associated radar data and ESM data are then weighted and fused.

2. The method for track fusion of a navigating vessel according to claim 1, characterized in that, The step of determining whether a data pair corresponds to the same target based on the similarity between data pairs includes: The overall distance similarity of the radar data pairs is determined based on the sum of the distance similarities between the radar data pairs at all time points; if the overall distance similarity between the radar data pairs is greater than or equal to the distance similarity threshold, the radar data pairs are determined to correspond to the same target. The overall carrier frequency similarity of the ESM data pairs is determined based on the sum of the carrier frequency similarities between the ESM data pairs at all time points; if the overall carrier frequency similarity between the ESM data pairs is greater than or equal to the carrier frequency similarity threshold, the ESM data pairs are determined to correspond to the same target. The overall azimuth similarity between the radar data and ESM data is determined based on the sum of the azimuth similarity between the radar data and ESM data at all time points. If the overall azimuth similarity between the radar and ESM sensors is greater than or equal to the azimuth similarity threshold, it is determined that the radar data and ESM data correspond to the same target.

3. The method for track fusion of a navigating vessel according to claim 2, characterized in that, The step of determining whether a data pair corresponds to the same target based on the similarity between data pairs includes: Adjusting the dynamic similarity threshold based on the monitoring time period to which the discrimination time point belongs includes: If the discrimination time point belongs to the initial monitoring stage, the dynamic similarity threshold is set as the first threshold. If the similarity is greater than or equal to the first threshold and the similarity is less than or equal to the sensor error, the data pair is determined to correspond to the same target. If the discrimination time point does not belong to the initial monitoring stage, the dynamic similarity threshold is set as the second threshold. If the similarity is greater than or equal to the second threshold, and among the similarities corresponding to time points before the discrimination time point, the similarities corresponding to time points with a preset proportion are all greater than or equal to the second threshold, then the data pair is determined to correspond to the same target.

4. The method for track fusion of a navigating vessel according to claim 3, characterized in that, The initial monitoring phase is determined based on the number of acquireable targets, which increases as the targets enter the monitoring range: If the number of targets that can be acquired increases to less than a preset threshold as the time it takes for the target to enter the monitoring range of the radar or ESM sensor increases, then the determination time point belongs to the initial monitoring stage. If the number of targets that can be acquired increases to a level greater than or equal to a preset threshold as the time it takes for the target to enter the monitoring range of the radar or ESM sensor increases, then the determination time point does not belong to the initial monitoring phase.

5. The method for track fusion of a navigating vessel according to claim 1, characterized in that, The method further includes: Before selecting radar data and ESM data for all targets at the discrimination time point, prior knowledge is acquired and used, including: If there is no temporal overlap between the radar data and the ESM data, then the radar data and the ESM data are excluded from the candidates for similarity calculation. If the similarity between the radar data and the ESM data has been obtained, then the similarity is cached. If the association between the radar data and ESM data has been determined, then the associated radar data and ESM data are cached. The radar data for each batch is matched one-to-one with each target.

6. The method for track fusion of a navigating vessel according to claim 1, characterized in that, The weighted fusion of associated radar data and ESM data includes: The associated radar data and ESM data are weighted and fused, and the time point of the ESM data is interpolated according to the time point of radar data acquisition. Singularities in the fused data are removed, and the fused data is then smoothed.

7. A trajectory fusion system for a navigating vessel, characterized in that, include: The data selection module is used to select radar data and ESM data of all targets before the judgment time point. All targets include all ships that have signals within the scanning period of the 2D radar or ESM sensor. The registration module is used to perform spatial and temporal registration of the radar data and ESM data of all the targets. The similarity calculation module is used to calculate the similarity between data pairs based on the nearest neighbor fast discrimination method. This includes filtering the data pairs that need to be similarity calculated from the radar data and ESM data of all targets; calculating the similarity between data pairs at all times before the discrimination time point, that is, calculating the distance similarity between radar data pairs, the carrier frequency similarity between ESM data pairs, and the azimuth similarity between radar data and ESM data at all times before the discrimination time point. The determination module is used to determine whether the data pairs correspond to the same target based on the similarity between the data pairs. This includes summing the similarity between the data pairs at all times before the determination time point to obtain the overall similarity between the data pairs; comparing the overall similarity between the data pairs with a dynamic similarity threshold, wherein the dynamic similarity threshold corresponds to the overall similarity between the data pairs, and the dynamic similarity threshold includes a distance similarity threshold, a carrier frequency similarity threshold, and a location similarity threshold. If the overall similarity between the data pairs is greater than or equal to the dynamic similarity threshold, the data pairs are determined to correspond to the same target. The association module is used to associate the data pairs if it is determined that the data pairs correspond to the same target; The fusion module is used to perform weighted fusion of the associated radar data and ESM data.

8. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by a processor, implements the steps of a track fusion method for a navigating vessel as described in any one of claims 1 to 7.

Images