A real-time fusion method and device of shipboard radar and AIS data

Abstract

The application relates to a real-time fusion method and device of shipborne radar and AIS data, and belongs to the technical field of data processing. The real-time fusion method of the shipborne radar and the AIS data comprises the following steps: acquiring shipborne radar data and AIS data of a target ship from a target time to a current time, and pre-processing the shipborne radar data; performing spatial alignment and time alignment on the AIS data and the pre-processed shipborne radar data, constructing a radar feature map based on the shipborne radar data subjected to the time alignment and the spatial alignment, and constructing an AIS feature map based on the AIS data subjected to the time alignment and the spatial alignment; and performing data fusion on the shipborne radar data and the AIS data subjected to the time alignment and the spatial alignment based on the radar feature map and the AIS feature map, so as to obtain a data fusion result. The application can effectively improve the precision and real-time performance of the fusion of the shipborne radar and the AIS data.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a method and apparatus for real-time fusion of shipborne radar and AIS data. Background Technology

[0002] Ship navigation safety is a crucial prerequisite for ensuring the high-quality development of waterway transportation. Due to the complexity and diversity of the marine meteorological environment, ship navigation safety heavily relies on shipborne sensors for real-time and accurate situational awareness of the constantly changing traffic environment. Shipborne radar and Automatic Identification Systems (AIS) are two core types of sensors for ship navigation; however, each type of sensor suffers from limitations in complex navigation environments, such as limited sensing range or uncertainties in the data collected.

[0003] Shipborne radar systems actively detect the distance, bearing, and speed of surrounding targets using electromagnetic waves, enabling all-weather operation. However, they are susceptible to interference from weather, sea conditions, and terrain obstacles, have blind spots, and cannot acquire static information about target vessels. AIS can provide high-precision static and dynamic data about ships, but it relies on Very High Frequency (VHF) communication, which significantly reduces the update frequency in open seas or in adverse weather conditions, and it cannot detect objects without AIS installed.

[0004] Radar systems provide real-time dynamic information about surrounding objects, while AIS provides static information and dynamic navigation status of ships. Effective fusion of these two technologies is crucial for achieving intelligent navigation. However, existing technologies for fusing shipborne radar and AIS data suffer from low accuracy and poor real-time performance. Summary of the Invention

[0005] In view of this, it is necessary to provide a real-time fusion method and apparatus for shipborne radar and AIS data to solve the problems of low accuracy and poor real-time performance in the fusion process of shipborne radar and AIS data in the existing technology.

[0006] To address the aforementioned problems, in a first aspect, the present invention provides a real-time fusion method for shipborne radar and AIS data, comprising:

[0007] Acquire the shipborne radar data and AIS data of the target vessel from the target time to the current time, and preprocess the shipborne radar data. The preprocessing of the shipborne radar data includes noise suppression processing and missing data reconstruction processing. The target time is the time corresponding to the preset time before the current time.

[0008] Spatially and temporally aligned AIS data and preprocessed shipborne radar data were performed. A radar feature map was constructed based on the temporally and spatially aligned shipborne radar data, and an AIS feature map was constructed based on the temporally and spatially aligned AIS data. The nodes of the radar feature map represent the position, heading angle, and speed information of each trajectory point in the shipborne radar data, and the edges of the radar feature map represent the distance difference and heading difference between any two trajectory points.

[0009] Data fusion is performed on time-aligned and space-aligned shipborne radar data and AIS data based on radar feature maps and AIS feature maps to obtain the data fusion result.

[0010] In one possible implementation, the data fusion of time-aligned and space-aligned shipborne radar data and AIS data based on radar feature maps and AIS feature maps to obtain the data fusion result includes:

[0011] A similarity matrix is ​​constructed based on radar feature maps and AIS feature maps. The matching results between shipborne radar data and AIS data after time and spatial alignment are determined based on the Hungarian algorithm and the similarity matrix. The diagonal elements of the similarity matrix are used to represent the similarity between nodes of radar feature maps and AIS feature maps, and the other elements of the similarity matrix are used to represent the similarity between edges of radar feature maps and AIS feature maps.

[0012] Based on the matching results, the set of trajectory points to be fused in the shipborne radar data and AIS data that have been time-aligned and spatially aligned is determined. Extended Kalman filtering is performed on the set of trajectory points to be fused, and statistical weighting is performed on the filtered set of trajectory points to be fused to obtain the data fusion result.

[0013] In one possible implementation, the preprocessing of the shipborne radar data includes:

[0014] The incremental DBSCAN algorithm is used to suppress noise in shipborne radar data, and the cubic spline interpolation algorithm is used to reconstruct missing data in shipborne radar data.

[0015] In one possible implementation, spatial alignment is performed on the AIS data and the preprocessed shipborne radar data, including:

[0016] The polar coordinate trajectory points in the preprocessed shipborne radar data relative to the shipborne radar origin are first transformed to a Cartesian coordinate system centered on the shipborne radar origin, and then the trajectory points in the Cartesian coordinate system are transformed to the WGS84 coordinate system based on the WGS84 coordinate information of the shipborne radar origin.

[0017] In one possible implementation, the transformation of the trajectory points converted to the Cartesian coordinate system to the WGS84 coordinate system includes:

[0018] The trajectory points converted to Cartesian coordinates are transformed to WGS84 coordinates using the following formula:

[0019]

[0020] Where point A is the origin of the shipborne radar, and point B is any trajectory point. This indicates the longitude of point B. Indicates the longitude of point A. This represents the azimuth angle of point B relative to point A. This represents the azimuth angle of point B. This represents the latitude of point B. This represents the latitude of point A. This represents the distance between point B and point A. This represents the Earth's radius.

[0021] In one possible implementation, time alignment is performed on the AIS data and the preprocessed shipborne radar data, including:

[0022] Time alignment of AIS data and preprocessed shipborne radar data is performed based on the following formula:

[0023]

[0024] in, This indicates that the shipborne radar is in the interpolation Longitude of time This indicates that the shipborne radar is sampling. Longitude of time This indicates that the shipborne radar is sampling. The speed at any moment, This indicates that the shipborne radar is sampling. The course of time This indicates that the shipborne radar is in the interpolation The dimension of time, This indicates that the shipborne radar is sampling. The dimension of time, This indicates that AIS is used in interpolation. Longitude of time Indicates that AIS is sampling Longitude of time Indicates that AIS is sampling The speed at any moment, Indicates that AIS is sampling The course of time This indicates that AIS is used in interpolation. The dimension of time, Indicates that AIS is sampling The latitude of time.

[0025] In one possible implementation, the statistical weighting operation on the filtered set of trajectory points to be fused includes:

[0026] The filtered set of trajectory points to be fused is statistically weighted based on the following formula:

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034]

[0035]

[0036]

[0037]

[0038]

[0039]

[0040]

[0041]

[0042]

[0043] in, This indicates the weighted and merged longitude. This represents the effective variance of the longitude observations from the shipborne radar. This represents the AIS longitude observation value. This represents the effective variance of AIS longitude observations. This represents the longitude observation value of the shipborne radar. This represents the variance of the longitude observations from the shipborne radar. This represents the variance of AIS longitude observations. This indicates a discrepancy between the longitude observations from the shipborne radar and the longitude observations from the AIS. This represents the dimensions after weighted fusion. This represents the effective variance of the latitude observations from the shipborne radar. Represents AIS latitude observation values. This represents the effective variance of AIS latitude observations. This represents the latitude observation value of the shipborne radar. This represents the variance of the latitude observations from the shipborne radar. This represents the variance of AIS latitude observations. This indicates a discrepancy between latitude observations from shipborne radar and latitude observations from AIS. This indicates the weighted and combined speed. This represents the effective variance of the shipborne radar speed observations. This represents the AIS airspeed observation value. This represents the effective variance of AIS airspeed observations. This represents the observed speed value from the ship's radar. This represents the variance of the observed ship speed values ​​from the ship's radar. This represents the variance of AIS airspeed observations. This indicates a discrepancy between shipborne radar speed observations and AIS speed observations. Indicates the course after weighted fusion. This represents the effective variance of the shipborne radar heading observations. This represents the AIS heading observation value. This represents the effective variance of the AIS heading observations. This represents the heading observation value of the shipborne radar. This represents the variance of the shipborne radar heading observations. This represents the variance of the AIS heading observations. This indicates a discrepancy between the shipborne radar heading observations and the AIS heading observations. , , , This represents the sensitivity adjustment coefficient, used to control the sensitivity of the effective variance to inconsistencies.

[0044] On the other hand, the present invention also provides a real-time fusion device for shipborne radar and AIS data, comprising:

[0045] The acquisition module is used to acquire the shipborne radar data and AIS data of the target vessel from the target time to the current time, and to preprocess the shipborne radar data. The preprocessing of the shipborne radar data includes noise suppression processing and missing data reconstruction processing. The target time is the time corresponding to the preset time before the current time.

[0046] The module is used to perform spatial and temporal alignment on AIS data and preprocessed shipborne radar data, and to construct radar feature maps based on the time-aligned and spatially aligned shipborne radar data, and AIS feature maps based on the time-aligned and spatially aligned AIS data. The nodes of the radar feature maps represent the position, heading angle and speed information of each trajectory point in the shipborne radar data, and the edges of the radar feature maps represent the distance difference and heading difference between any two trajectory points.

[0047] The fusion module is used to fuse time-aligned and space-aligned shipborne radar data and AIS data based on radar feature maps and AIS feature maps to obtain the data fusion result.

[0048] Secondly, the present invention also provides a data fusion device, including a memory and a processor, wherein,

[0049] The memory is used to store programs;

[0050] The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the real-time fusion method of shipborne radar and AIS data described in any of the above implementations.

[0051] Thirdly, the present invention also provides a computer-readable storage medium for storing a computer-readable program or instruction, which, when executed by a processor, can implement the steps in the real-time fusion method of shipborne radar and AIS data described in any of the above implementations.

[0052] The beneficial effects of this invention are as follows: The real-time fusion method and apparatus for shipborne radar and AIS data provided by this invention firstly ensures the accuracy and reliability of the shipborne radar data by performing noise suppression processing and missing data reconstruction processing, thereby ensuring the accuracy of subsequent data fusion. Next, spatial and temporal alignment is performed on the AIS data and shipborne radar data to provide a foundation for subsequent data fusion. Then, a feature map is constructed to improve the efficiency of data association, thereby improving the real-time performance of data fusion. Finally, the data fusion result is obtained through the constructed feature map. This invention can effectively improve the accuracy and real-time performance of shipborne radar and AIS data fusion. Attached Figure Description

[0053] Figure 1 A schematic flowchart of an embodiment of the real-time fusion method of shipborne radar and AIS data provided by the present invention;

[0054] Figure 2 A schematic flowchart of an embodiment of the overall process of real-time data fusion provided by the present invention;

[0055] Figure 3 A schematic flowchart of an embodiment of the noise point screening process provided by the present invention;

[0056] Figure 4 A schematic flowchart of an embodiment of the ship trajectory reconstruction process provided by the present invention;

[0057] Figure 5 A schematic flowchart of an embodiment of the data spatiotemporal alignment process provided by the present invention;

[0058] Figure 6 A schematic flowchart of an embodiment of the data matching process provided by the present invention;

[0059] Figure 7 A schematic flowchart of an embodiment of the data fusion process provided by the present invention;

[0060] Figure 8 A schematic diagram of an embodiment of the real-time fusion device for shipborne radar and AIS data provided by the present invention;

[0061] Figure 9 A schematic diagram of an embodiment of the data fusion device provided by the present invention. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0063] In the description of the embodiments of the present invention, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.

[0064] The terms "first," "second," etc., used in the embodiments of this invention are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a technical feature defined with "first" or "second" may explicitly or implicitly include at least one of that feature.

[0065] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0066] This invention provides a real-time fusion method and apparatus for shipborne radar and AIS data, which will be described below.

[0067] Figure 1 A schematic flowchart of an embodiment of the real-time fusion method for shipborne radar and AIS data provided by the present invention is shown below. Figure 1 As shown, the real-time fusion method of shipborne radar and AIS data includes:

[0068] S101. Acquire the shipborne radar data and AIS data of the target vessel from the target time to the current time, and preprocess the shipborne radar data. The preprocessing of the shipborne radar data includes noise suppression processing and missing data reconstruction processing. The target time is the time corresponding to the preset time before the current time.

[0069] It should be noted that by performing noise suppression and missing data reconstruction on shipborne radar data, the accuracy and reliability of the data can be guaranteed, the motion characteristics of the trajectory can be restored, and the accuracy of subsequent data fusion can be improved.

[0070] S102. Spatially and temporally align the AIS data and the preprocessed shipborne radar data, and construct a radar feature map based on the temporally and spatially aligned shipborne radar data. Construct an AIS feature map based on the temporally and spatially aligned AIS data. The nodes of the radar feature map represent the position, heading angle, and speed information of each trajectory point in the shipborne radar data, and the edges of the radar feature map represent the distance difference and heading difference between any two trajectory points.

[0071] It should be noted that because shipborne radar and AIS describe the position of ship targets differently, and their data time intervals also differ significantly, spatial and temporal alignment of AIS and shipborne radar data is necessary to provide a foundation for subsequent data fusion. Constructing radar feature maps and AIS feature maps based on AIS and shipborne radar data allows for more precise data correlation, thereby improving the accuracy of subsequent data fusion.

[0072] S103. Based on radar feature maps and AIS feature maps, perform data fusion on time-aligned and space-aligned shipborne radar data and AIS data to obtain data fusion results.

[0073] It should be noted that when data is correlated using radar feature maps and AIS feature maps, the efficiency of data correlation can be effectively improved, thereby enhancing the real-time performance of data fusion.

[0074] In summary, the real-time fusion method for shipborne radar and AIS data provided in this invention first performs noise suppression and missing data reconstruction on the shipborne radar data to ensure the accuracy and reliability of the data, thereby ensuring the accuracy of subsequent data fusion. Next, spatial and temporal alignment is performed on the AIS data and shipborne radar data to provide a foundation for subsequent data fusion. Then, a feature map is constructed to improve the efficiency of data association, thereby improving the real-time performance of data fusion. Finally, the data fusion result is obtained through the constructed feature map. This invention can effectively improve the accuracy and real-time performance of shipborne radar and AIS data fusion.

[0075] Combination Figure 2 The overall process of real-time data fusion provided by this invention includes the following steps:

[0076] 1. Noise suppression and missing data reconstruction.

[0077] In some embodiments of the present invention, the preprocessing of shipborne radar data includes:

[0078] The incremental DBSCAN algorithm is used to suppress noise in shipborne radar data, and the cubic spline interpolation algorithm is used to reconstruct missing data in shipborne radar data.

[0079] The incremental DBSCAN algorithm is used to identify and remove discrete noise data to ensure the accuracy and reliability of the data; the cubic spline interpolation algorithm is used to reconstruct the trajectory of the missing parts to restore the motion characteristics of the trajectory.

[0080] 1) Incremental DBSCAN algorithm to remove noisy data.

[0081] Combination Figure 3 The noise point screening process is as follows:

[0082] a. Set the neighborhood radius parameter and the minimum number of members in the cluster.

[0083] b. For newly arrived point P in the data stream, perform neighborhood query and online status update.

[0084] c. Remove expired points Q in the data stream and maintain the cluster structure.

[0085] d. Points that are continuously marked as noise and are not absorbed by any cluster are suppressed noise points.

[0086] 2) The cubic spline interpolation algorithm is used to reconstruct noisy data.

[0087] Combination Figure 4 The data reconstruction process is as follows:

[0088] a. Input trajectory data.

[0089] b. Construct a cubic spline interpolation function.

[0090] c. Set function boundary conditions.

[0091] d. Solve for the interpolation parameters.

[0092] e. Calculate the interpolation points and add them to the trajectory.

[0093] f. Output the complete trajectory.

[0094] 2. Spatiotemporal alignment.

[0095] Combination Figure 5 As can be seen, shipborne radar and AIS sensors describe the position of ship targets using different methods, and the time intervals of their data also differ significantly. Therefore, data from these two different reference frames cannot be directly correlated and fused. To address the temporal and spatial asymmetry between the two types of data, spatial and temporal alignment is necessary.

[0096] 1) Spatial alignment.

[0097] The standard coordinate system used for AIS data is the WGS-84 coordinate system, which uses the center of the Earth as the origin and longitude, latitude, and altitude to describe the location on Earth.

[0098] Shipborne radar uses a ship coordinate system, with its origin located at the ship's center, representing the position and distance of objects around the ship. Essentially, it is a polar coordinate system; the distance to the pole is called the radial coordinate, and the angle is called the angular coordinate.

[0099] In some embodiments of the present invention, spatial alignment of AIS data and preprocessed shipborne radar data includes:

[0100] The polar coordinate trajectory points in the preprocessed shipborne radar data relative to the shipborne radar origin are first transformed to a Cartesian coordinate system centered on the shipborne radar origin, and then the trajectory points in the Cartesian coordinate system are transformed to the WGS84 coordinate system based on the WGS84 coordinate information of the shipborne radar origin.

[0101] In some embodiments of the present invention, the step of converting the trajectory points transformed to the Cartesian coordinate system to the WGS84 coordinate system includes:

[0102] The trajectory points converted to Cartesian coordinates are transformed to WGS84 coordinates using the following formula:

[0103]

[0104] Where point A is the origin of the shipborne radar, and point B is any trajectory point. This indicates the longitude of point B. Indicates the longitude of point A. This represents the angular distance between point B and point A. This represents the azimuth angle of point B. This represents the latitude of point B. This represents the latitude of point A. This represents the distance between point B and point A. This represents the Earth's radius.

[0105] 2) Time alignment.

[0106] Shipborne radar and AIS have different data update frequencies. Radar scans one revolution every 3 seconds, while AIS updates dynamic data every 3-180 seconds. This invention uses an interpolation-extrapolation method to solve the data time alignment problem. The ship information at the interpolated time is calculated based on the information from the previous and next moments of the sensor target.

[0107] In some embodiments of the present invention, time alignment is performed on AIS data and preprocessed shipborne radar data, including:

[0108] Time alignment of AIS data and preprocessed shipborne radar data is performed based on the following formula:

[0109]

[0110] in, This indicates that the shipborne radar is in the interpolation Longitude of time This indicates that the shipborne radar is sampling. Longitude of time This indicates that the shipborne radar is sampling. The speed at any moment, This indicates that the shipborne radar is sampling. The course of time This indicates that the shipborne radar is in the interpolation The dimension of time, This indicates that the shipborne radar is sampling. The dimension of time, This indicates that AIS is used in interpolation. Longitude of time Indicates that AIS is sampling Longitude of time Indicates that AIS is sampling The speed at any moment, Indicates that AIS is sampling The course of time This indicates that AIS is used in interpolation. The dimension of time, Indicates that AIS is sampling The latitude of time.

[0111] 3. Data association.

[0112] Combination Figure 6 Here are the specific steps involved in data association and matching:

[0113] 1) Construct feature maps of AIS and radar data.

[0114] Define two feature maps that represent AIS and radar data, respectively.

[0115] 2) Define a similarity matrix.

[0116] In some embodiments of the present invention, the data fusion of time-aligned and space-aligned shipborne radar data and AIS data based on radar feature maps and AIS feature maps to obtain data fusion results includes:

[0117] A similarity matrix is ​​constructed based on radar feature maps and AIS feature maps. The matching results between shipborne radar data and AIS data after time and spatial alignment are determined based on the Hungarian algorithm and the similarity matrix. The diagonal elements of the similarity matrix are used to represent the similarity between nodes of radar feature maps and AIS feature maps, and the other elements of the similarity matrix are used to represent the similarity between edges of radar feature maps and AIS feature maps.

[0118] Based on the matching results, the set of trajectory points to be fused in the shipborne radar data and AIS data that have been time-aligned and spatially aligned is determined. Extended Kalman filtering is performed on the set of trajectory points to be fused, and statistical weighting is performed on the filtered set of trajectory points to be fused to obtain the data fusion result.

[0119] 3) Use the permutation matrix to find the best matching data from the similarity matrix.

[0120] 4) The matching result is obtained by finding the optimal similarity matrix through the Hungarian algorithm.

[0121] 4. Data fusion.

[0122] Considering the nonlinear characteristics of ship navigation status, an extended Kalman filter model adapted to the ship dynamic model is established to achieve accurate estimation of ship navigation status, and then statistical weighted fusion is performed to obtain fused data.

[0123] Combination Figure 7 The specific process of data fusion includes:

[0124] 1) Extended Kalman Filter Algorithm.

[0125] Initialization: Set the initial state estimate and the initial covariance matrix to characterize the uncertainty of the state estimate.

[0126] Prediction steps:

[0127] State prediction: Predicting the state at the next time step using a nonlinear state transition function.

[0128] Covariance prediction: Calculate the Jacobian matrix of the state transition function.

[0129] Update steps:

[0130] Observation prediction: Predicting observed values ​​using a nonlinear observation function.

[0131] Calculation of observation residuals: Calculate the residuals based on the actual observed values.

[0132] Linearized observation model: Calculate the Jacobian matrix of the observation function.

[0133] Kalman gain calculation: Calculate the gain based on the predicted covariance value and the observation noise covariance matrix.

[0134] State update: Correct the state estimate using Kalman gain.

[0135] Covariance Update: Update the covariance matrix to reflect the corrected uncertainty.

[0136] Iterative loop: Repeat the prediction and update steps until the state estimates for all time points are completed.

[0137] By applying extended Kalman filtering to estimate the longitude, latitude, speed, and heading of the target ship, we can obtain estimation results with smaller measurement error variance and closer to the true value of the track.

[0138] 2) Statistical weighting algorithm.

[0139] In some embodiments of the present invention, the step of performing statistical weighting operations on the filtered set of trajectory points to be fused includes:

[0140] The filtered set of trajectory points to be fused is statistically weighted based on the following formula:

[0141]

[0142]

[0143]

[0144]

[0145]

[0146]

[0147]

[0148]

[0149]

[0150]

[0151]

[0152]

[0153]

[0154]

[0155]

[0156]

[0157] in, This indicates the weighted and merged longitude. This represents the effective variance of the longitude observations from the shipborne radar. This represents the AIS longitude observation value. This represents the effective variance of AIS longitude observations. This represents the longitude observation value of the shipborne radar. This represents the variance of the longitude observations from the shipborne radar. This represents the variance of AIS longitude observations. This indicates a discrepancy between the longitude observations from the shipborne radar and the longitude observations from the AIS. This represents the dimensions after weighted fusion. This represents the effective variance of the latitude observations from the shipborne radar. Represents AIS latitude observation values. This represents the effective variance of AIS latitude observations. This represents the latitude observation value of the shipborne radar. This represents the variance of the latitude observations from the shipborne radar. This represents the variance of AIS latitude observations. This indicates a discrepancy between latitude observations from shipborne radar and latitude observations from AIS. This indicates the weighted and combined speed. This represents the effective variance of the shipborne radar speed observations. This represents the AIS airspeed observation value. This represents the effective variance of AIS airspeed observations. This represents the observed speed value from the ship's radar. This represents the variance of the observed ship speed values ​​from the ship's radar. This represents the variance of AIS airspeed observations. This indicates a discrepancy between shipborne radar speed observations and AIS speed observations. Indicates the course after weighted fusion. This represents the effective variance of the shipborne radar heading observations. This represents the AIS heading observation value. This represents the effective variance of the AIS heading observations. This represents the heading observation value of the shipborne radar. This represents the variance of the shipborne radar heading observations. This represents the variance of the AIS heading observations. This indicates a discrepancy between the shipborne radar heading observations and the AIS heading observations. , , , This represents the sensitivity adjustment coefficient, used to control the sensitivity of the effective variance to inconsistencies.

[0158] The location information in AIS data is transmitted by the Global Positioning System, which has high accuracy; therefore, the variance of each AIS observation can be set to 0.00000001. The variance of each observation from shipborne radar is determined by its measurement accuracy. For example, if the radar distance measurement accuracy is 0.9%, the radar distance variance is 0.000081; if the angle measurement error is 1°, the maximum radar angle measurement accuracy is 0.28%, and the radar azimuth error is 0.0000077. The sensitivity adjustment coefficient is a positive constant used to control the sensitivity of inconsistencies to effective variance adjustment. Different sensitivity adjustment coefficient values ​​can be used for longitude, latitude, SOG, and COG.

[0159] In this way, when significant discrepancies arise between AIS and radar observations, leading to increased inconsistency, the effective variance of both will increase accordingly. Based on the principle of weighted averaging, the weights are proportional to the inverse of the effective variance. Data sources whose effective variance is significantly amplified due to inconsistencies with other sensors will automatically have their weights reduced in the current fusion step. This allows the fusion results to adaptively mitigate the negative impact of potential outlier data or short-term sensor performance degradation, favoring data with higher consistency. Consequently, more accurate data fusion results can be obtained in complex and non-ideal ship navigation environments.

[0160] Extended Kalman filtering is applied to radar and AIS data, followed by statistical weighting of the filtered data. The aim is to further eliminate measurement noise and uncertainty through continuous track point processing, resulting in smoother and more accurate target state estimation.

[0161] This invention addresses the uncertainties inherent in shipborne radar and AIS data in complex waters by proposing preprocessing techniques to improve the quality of both data. To address the inconsistency in positional accuracy between shipborne radar and AIS data, a feature map matching-based method for associating radar and AIS data is proposed, achieving high-precision association. Furthermore, to meet the real-time requirements of shipborne radar and AIS data fusion, an extended Kalman filter and statistical weighting algorithm are proposed to accurately predict target vessel trajectories, balancing the accuracy and efficiency of data fusion.

[0162] This invention assesses the effectiveness of continuous data association by setting simulated noise. A traditional multi-threshold association algorithm was used as a comparative experiment, as shown in Table 1. When the noise level is 0.00001, the traditional multi-threshold association algorithm performs reasonably well. However, as the noise increases, when the noise level is 0.0001, the traditional method can only associate 18 matching pairs, with one pair not being associated. When the noise level continues to increase while the threshold remains unchanged, the traditional method can only associate 9 matching pairs, with three incorrect associations, resulting in a matching success rate of 66.7%. In terms of computational time, the traditional method has a smaller computational load and faster computation speed.

[0163] Table 1: Continuous Data Simulation Results

[0164]

[0165] This invention also uses simulated noise to determine the effectiveness of sparse data association. A traditional multi-threshold association algorithm was used as a comparative experiment, as shown in Table 2. When the noise level was 0.0001, the traditional multi-threshold association algorithm performed reasonably well. However, as the noise gradually increased while the threshold remained unchanged, more points failed to be associated and matched.

[0166] Table 2: Simulation Results for Sparse Data

[0167]

[0168] The data association method provided by this invention achieves significantly higher matching accuracy than traditional algorithms. In terms of processing time, the proposed matching method takes far less than 3 seconds, making it capable of real-time association of AIS and radar information.

[0169] Furthermore, the present invention has also conducted experiments to verify the denoising performance, accuracy, and data processing time of the data fusion method provided by the present invention, thus confirming the practicality of the data fusion method provided by the present invention.

[0170] To better implement the real-time fusion method of shipborne radar and AIS data in the embodiments of the present invention, based on the real-time fusion method of shipborne radar and AIS data, correspondingly, as follows: Figure 8 As shown, this embodiment of the invention also provides a real-time fusion device for shipborne radar and AIS data. The real-time fusion device 800 for shipborne radar and AIS data includes:

[0171] The acquisition module 801 is used to acquire the shipborne radar data and AIS data of the target ship from the target time to the current time, and to preprocess the shipborne radar data. The preprocessing of the shipborne radar data includes noise suppression processing and missing data reconstruction processing. The target time is the time corresponding to the preset time before the current time.

[0172] Module 802 is used to perform spatial and temporal alignment on AIS data and preprocessed shipborne radar data, and to construct radar feature maps based on the time-aligned and spatially aligned shipborne radar data, and AIS feature maps based on the time-aligned and spatially aligned AIS data. The nodes of the radar feature maps are used to represent the position, heading angle and speed information of each trajectory point in the shipborne radar data, and the edges of the radar feature maps are used to represent the distance difference and heading difference between any two trajectory points. The nodes of the AIS feature maps are used to represent the position, heading angle and speed information of each trajectory point in the AIS data, and the edges of the AIS feature maps are used to represent the distance difference and heading difference between any two trajectory points.

[0173] The fusion module 803 is used to perform data fusion on time-aligned and space-aligned shipborne radar data and AIS data based on radar feature maps and AIS feature maps to obtain data fusion results.

[0174] The real-time fusion device 800 for shipborne radar and AIS data provided in the above embodiments can realize the technical solutions described in the above embodiments of the real-time fusion method for shipborne radar and AIS data. The specific implementation principles of each module or unit can be found in the corresponding content in the above embodiments of the real-time fusion method for shipborne radar and AIS data, which will not be repeated here.

[0175] like Figure 9 As shown, the present invention also provides a data fusion device 900. The data fusion device 900 includes a processor 901, a memory 902, and a display 903. Figure 9 Only some components of the data fusion device 900 are shown; however, it should be understood that implementation of all shown components is not required, and more or fewer components may be implemented instead.

[0176] In some embodiments, processor 901 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 902 or process data, such as the magnetic resonance image optimization method of the present invention.

[0177] In some embodiments, processor 901 may be a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, processor 901 may be local or remote. In some embodiments, processor 901 may be implemented on a cloud platform. In one embodiment, the cloud platform may include a private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-cloud, etc., or any combination thereof.

[0178] In some embodiments, memory 902 may be an internal storage unit of the data fusion device 900, such as a hard disk or memory of the data fusion device 900. In other embodiments, memory 902 may also be an external storage device of the data fusion device 900, such as a pluggable hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the data fusion device 900.

[0179] Furthermore, the memory 902 may include both internal storage units of the data fusion device 900 and external storage devices. The memory 902 is used to store application software and various types of data for which the data fusion device 900 is installed.

[0180] In some embodiments, display 903 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an organic light-emitting diode (OLED) touchscreen. Display 903 is used to display information from the data fusion device 900 and to display a visual user interface. Components 901-903 of the data fusion device 900 communicate with each other via a system bus.

[0181] In one embodiment, when the processor 901 executes the real-time fusion program of shipborne radar and AIS data in the memory 902, the following steps can be implemented:

[0182] Acquire the shipborne radar data and AIS data of the target vessel from the target time to the current time, and preprocess the shipborne radar data. The preprocessing of the shipborne radar data includes noise suppression processing and missing data reconstruction processing. The target time is the time corresponding to the preset time before the current time.

[0183] Spatially and temporally aligned AIS data and preprocessed shipborne radar data were performed. A radar feature map was constructed based on the temporally and spatially aligned shipborne radar data, and an AIS feature map was constructed based on the temporally and spatially aligned AIS data. The nodes of the radar feature map represent the position, heading angle, and speed information of each trajectory point in the shipborne radar data, and the edges of the radar feature map represent the distance difference and heading difference between any two trajectory points.

[0184] Data fusion is performed on time-aligned and space-aligned shipborne radar data and AIS data based on radar feature maps and AIS feature maps to obtain the data fusion result.

[0185] It should be understood that when the processor 901 executes the real-time fusion program of shipborne radar and AIS data in the memory 902, in addition to the functions mentioned above, it can also perform other functions, as can be found in the description of the corresponding method embodiments above.

[0186] Furthermore, this embodiment of the invention does not specifically limit the type of the data fusion device 900 mentioned. The data fusion device 900 can be a portable electronic device such as a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, or laptop computer. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic devices can also be other portable electronic devices, such as laptop computers with touch-sensitive surfaces (e.g., touch panels). It should also be understood that in some other embodiments of the invention, the data fusion device 900 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).

[0187] Accordingly, this application also provides a computer-readable storage medium for storing computer-readable programs or instructions. When the programs or instructions are executed by a processor, they can implement the steps or functions of the real-time fusion method of shipborne radar and AIS data provided in the above-described method embodiments.

[0188] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware (such as a processor, controller, etc.), and the computer program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0189] The above provides a detailed description of the real-time fusion method and apparatus for shipborne radar and AIS data provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A real-time fusion method for shipborne radar and AIS data, characterized in that, include: Acquire the shipborne radar data and AIS data of the target vessel from the target time to the current time, and preprocess the shipborne radar data. The preprocessing of the shipborne radar data includes noise suppression processing and missing data reconstruction processing. The target time is the time corresponding to the preset time before the current time. Spatially and temporally aligned AIS data and preprocessed shipborne radar data were performed. A radar feature map was constructed based on the temporally and spatially aligned shipborne radar data, and an AIS feature map was constructed based on the temporally and spatially aligned AIS data. The nodes of the radar feature map represent the position, heading angle, and speed information of each trajectory point in the shipborne radar data, and the edges of the radar feature map represent the distance difference and heading difference between any two trajectory points. Based on radar feature maps and AIS feature maps, time-aligned and space-aligned shipborne radar data and AIS data are fused to obtain the data fusion result; The data fusion is performed on time-aligned and space-aligned shipborne radar data and AIS data based on radar feature maps and AIS feature maps to obtain the data fusion result, including: A similarity matrix is ​​constructed based on radar feature maps and AIS feature maps. The matching results between shipborne radar data and AIS data after time and spatial alignment are determined based on the Hungarian algorithm and the similarity matrix. The diagonal elements of the similarity matrix are used to represent the similarity between nodes of radar feature maps and AIS feature maps, and the other elements of the similarity matrix are used to represent the similarity between edges of radar feature maps and AIS feature maps. Based on the matching results, the set of trajectory points to be fused in the shipborne radar data and AIS data that have been time-aligned and spatially aligned is determined. Extended Kalman filtering is performed on the set of trajectory points to be fused, and statistical weighting is performed on the filtered set of trajectory points to be fused to obtain the data fusion result.

2. The real-time fusion method of shipborne radar and AIS data according to claim 1, characterized in that, The preprocessing of shipborne radar data includes: The incremental DBSCAN algorithm is used to suppress noise in shipborne radar data, and the cubic spline interpolation algorithm is used to reconstruct missing data in shipborne radar data.

3. The real-time fusion method of shipborne radar and AIS data according to claim 1, characterized in that, Spatial alignment of AIS data and preprocessed shipborne radar data, including: The polar coordinate trajectory points in the preprocessed shipborne radar data relative to the shipborne radar origin are first transformed to a Cartesian coordinate system centered on the shipborne radar origin, and then the trajectory points in the Cartesian coordinate system are transformed to the WGS84 coordinate system based on the WGS84 coordinate information of the shipborne radar origin.

4. The real-time fusion method of shipborne radar and AIS data according to claim 3, characterized in that, Transform the trajectory points converted to the Cartesian coordinate system to the WGS84 coordinate system, including: The trajectory points converted to Cartesian coordinates are transformed to WGS84 coordinates using the following formula: Where point A is the origin of the shipborne radar, and point B is any trajectory point. This indicates the longitude of point B. Indicates the longitude of point A. This represents the azimuth angle of point B relative to point A. This represents the azimuth angle of point B. This represents the latitude of point B. This represents the latitude of point A. This represents the distance between point B and point A. This represents the Earth's radius.

5. The real-time fusion method of shipborne radar and AIS data according to claim 1, characterized in that, Time alignment of AIS data and preprocessed shipborne radar data, including: Time alignment of AIS data and preprocessed shipborne radar data is performed based on the following formula: in, This indicates that the shipborne radar is in the interpolation Longitude of time This indicates that the shipborne radar is sampling. Longitude of time This indicates that the shipborne radar is sampling. The speed at any moment, This indicates that the shipborne radar is sampling. The course of time This indicates that the shipborne radar is in the interpolation The dimension of time, This indicates that the shipborne radar is sampling. The dimension of time, This indicates that AIS is used in interpolation. Longitude of time Indicates that AIS is sampling Longitude of time Indicates that AIS is sampling The speed at any moment, Indicates that AIS is sampling The course of time This indicates that AIS is used in interpolation. The dimension of time, Indicates that AIS is sampling The latitude of time.

6. The real-time fusion method of shipborne radar and AIS data according to claim 1, characterized in that, The step of performing statistical weighting operations on the filtered set of trajectory points to be fused includes: The filtered set of trajectory points to be fused is statistically weighted based on the following formula: in, This indicates the weighted and merged longitude. This represents the effective variance of the longitude observations from the shipborne radar. This represents the AIS longitude observation value. This represents the effective variance of AIS longitude observations. This represents the longitude observation value of the shipborne radar. This represents the variance of the longitude observations from the shipborne radar. This represents the variance of AIS longitude observations. This indicates a discrepancy between the longitude observations from the shipborne radar and the longitude observations from the AIS. This represents the dimensions after weighted fusion. This represents the effective variance of the latitude observations from the shipborne radar. Represents AIS latitude observation values. This represents the effective variance of AIS latitude observations. This represents the latitude observation value of the shipborne radar. This represents the variance of the latitude observations from the shipborne radar. This represents the variance of AIS latitude observations. This indicates a discrepancy between latitude observations from shipborne radar and latitude observations from AIS. This indicates the weighted and combined speed. This represents the effective variance of the shipborne radar speed observations. This represents the AIS airspeed observation value. This represents the effective variance of AIS airspeed observations. This represents the observed speed value from the ship's radar. This represents the variance of the observed ship speed values ​​from the ship's radar. This represents the variance of AIS airspeed observations. This indicates a discrepancy between shipborne radar speed observations and AIS speed observations. Indicates the course after weighted fusion. This represents the effective variance of the shipborne radar heading observations. This represents the AIS heading observation value. This represents the effective variance of the AIS heading observations. This represents the heading observation value of the shipborne radar. This represents the variance of the shipborne radar heading observations. This represents the variance of the AIS heading observations. This indicates a discrepancy between the shipborne radar heading observations and the AIS heading observations. , , , This represents the sensitivity adjustment coefficient, used to control the sensitivity of the effective variance to inconsistencies.

7. A real-time fusion device for shipborne radar and AIS data, characterized in that, include: The acquisition module is used to acquire the shipborne radar data and AIS data of the target vessel from the target time to the current time, and to preprocess the shipborne radar data. The preprocessing of the shipborne radar data includes noise suppression processing and missing data reconstruction processing. The target time is the time corresponding to the preset time before the current time. The module is used to perform spatial and temporal alignment on AIS data and preprocessed shipborne radar data, and to construct radar feature maps based on the time-aligned and spatially aligned shipborne radar data, and AIS feature maps based on the time-aligned and spatially aligned AIS data. The nodes of the radar feature maps represent the position, heading angle and speed information of each trajectory point in the shipborne radar data, and the edges of the radar feature maps represent the distance difference and heading difference between any two trajectory points. The fusion module is used to fuse time-aligned and space-aligned shipborne radar data and AIS data based on radar feature maps and AIS feature maps to obtain the data fusion result; The data fusion is performed on time-aligned and space-aligned shipborne radar data and AIS data based on radar feature maps and AIS feature maps to obtain the data fusion result, including: A similarity matrix is ​​constructed based on radar feature maps and AIS feature maps. The matching results between shipborne radar data and AIS data after time and spatial alignment are determined based on the Hungarian algorithm and the similarity matrix. The diagonal elements of the similarity matrix are used to represent the similarity between nodes of radar feature maps and AIS feature maps, and the other elements of the similarity matrix are used to represent the similarity between edges of radar feature maps and AIS feature maps. Based on the matching results, the set of trajectory points to be fused in the shipborne radar data and AIS data that have been time-aligned and spatially aligned is determined. Extended Kalman filtering is performed on the set of trajectory points to be fused, and statistical weighting is performed on the filtered set of trajectory points to be fused to obtain the data fusion result.

8. A data fusion device, characterized in that, Including memory and processor, among which, The memory is used to store programs; The processor, coupled to the memory, is used to execute the program stored in the memory to implement the steps in the real-time fusion method of shipborne radar and AIS data according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, Used to store computer-readable programs or instructions, which, when executed by a processor, can implement the steps in the real-time fusion method of shipborne radar and AIS data as described in any one of claims 1 to 6.

Images